THE WORLD’S LEADING INTERNATIONAL CONFERENCE & EXHIBITION FOR DESIGN, DEVELOPMENT & TESTING OF AUTONOMOUS VEHICLES

5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY

10% EARLY-BOOKING DISCOUNT! Go online to book your pass!

LATEST UPDATES ANNOUNCED! MORE REASONS TO ATTEND THAN EVER BEFORE OUR BIGGEST EVENT TO DATE!

An integral part of Pass includes Nvidia 150+ speakers & Siemens workshops Autonomous Vehicle Over 500 delegates TECHNOLOGY expected WORLD EXPO 2018 Autonomous Vehicle Test & Development Symposium

Event sponsor

10% DISCOUNT*

EARLY-BOOKINGGovernment organisations DISCOUNT! Welcome receive a 20% We are fast approaching the biggest autonomous vehicle event in Europe, taking place this June in Stuttgart. There will be 150+ expert speakers and over 500 delegates taking part across three conferences, and 80+ leading AV technology companies exhibiting. We also have two exclusive workshops from Nvidia and Siemens taking place on 7 June, both of which are included in the delegate pass fee. You can read more about the workshops on page 21. This programme is divided into three distinct sections – one for each symposium. Please take time to read about each 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY speaker and their presentation synopsis, to better understand what can be learned and to realise the great networking THE WORLD’S ONLY INTERNATIONAL CONFERENCE DEDICATED TO TESTING AND opportunities that can be gained by meeting the speakers. For 2018, we have created a ‘one pass, access all areas’ for delegates. This means that your pass will grant you access to VALIDATION OF AUTONOMOUS VEHICLES AND SELF-DRIVING TECHNOLOGY the Autonomous Vehicle Test & Development Symposium, Autonomous Vehicle Interior Design & Technology Symposium, Autonomous Vehicle Software Symposium, the Nvidia and Siemens workshops, plus the Autonomous Vehicle Technology World Expo. Of course, your refreshments and lunch will be provided across all three days, and we invite you to attend the 〉〉〉 pre-event opening networking breakfast on Day 1 and the evening drinks reception on Wednesday 6 June. UPDATED PROGRAMME We look forward to welcoming you to Stuttgart on 5, 6 and 7 June.

Best regards, Andrew Boakes, conference director 20% *DISCOUNT FOR GOVERNMENT ORGANISATIONS!

PAGE 1 PAGE 12 PAGE 17

*OVER 150 SPEAKERS • MORE THAN 500 DELEGATES ATTENDED IN 2017 • ATTENDEES VISITED FROM OVER 40 COUNTRIES IN 2017 • HELD ALONGSIDE AUTONOMOUS VEHICLE TECHNOLOGY WORLD EXPO – MEET OVER 80 EXHIBITORS! ONE PASS GIVES ACCESS TO ALL EVENTS AT NO EXTRA COST!

MORE STREAMS CONFERENCE PASS RATES & PACKAGE ADDED FOR • Entry to ALL three symposiums Group Bookings 2018 – OUR • Access to the proceedings of those conferences attended Receive an extra 10% discount on each registration for BIGGEST EVENT • Networking drinks reception a group booking (2+ delegates) by making them on the YET! • VIP delegate and speaker dining same date, from the same company. • Entry to Nvidia and Siemens workshops Government organisations receive a 20% discount *Contact: [email protected] An integral part of

www.autonomousvehiclesymposium.com 3-DAY PASS...... €1,950 €1,755 + VAT Autonomous Vehicle www.autonomousvehicleinteriors.com TECHNOLOGY 2-DAY PASS...... €1,750 €1,575 + VAT www.autonomousvehicle-software.com WORLD EXPO 2018

* Expected numbers www.autonomousvehiclesymposium.com 1 Autonomous Vehicle Test & Development Symposium

10% EARLY-BIRD DISCOUNT DAY 1 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 1

DAY 1 TUESDAY 5 JUNE need to be driven for over 11 billion miles. However, novel systems engineering method – an extension (pods). One of the features of the proposed method is rather than number of miles it is the quality of miles to the STPA method – has been developed and the ability to create test scenarios and pass criteria. that is important. To find the ‘interesting scenarios’, a applied to low-speed automated driving systems

09:00 - 12:45 - Room A - Keynote Presentations and Conference Opening 14:15 - 18:00 - Room B - Vision – Sensors, Lidar & Mapping Technology

09:00 - Test development and to look-ahead road topology and world 10:30 - 11:15 - Break 14:15 - Maps as a sensor for air with map data. Within OADF, several standards like and integrated lidar positioning. The use of lidar as execution system for autonomously model attributes using HD maps. the autonomous car NDS, ADASIS, SENSORIS and TISA will cooperate. a reference or complementary system for camera- performed scenarios 11:15 - Validation of L4 urban Mike Tzamaloukas, vice president, based or sensor-fused detection will also be addressed Mathias Klumpf, test and development 10:00 - Insuring autonomous (test) driving automated driving BU autonomous drive/ADAS and 15:15 - Enabling autonomous with regard to the lidar vs. camera positioning. engineer, Audi AG, Germany Andreas Bradt, project manager autonomous Dr Christian Krummel, VP, Robert CoC navigation, Harman, Germany vehicles in inclement weather To ensure highest quality and safety, complex testing driving, Allianz Automotive, Germany Bosch GmbH, Germany With the advent of the autonomous car, modelling the Phil Magney, founder and principal, 17:00 - Determining precise boresight of autonomous driving assistance is performed with After more 130 years of automotive development, Validating a fully automated vehicle operating environment has been a challenge and an opportunity VSI Labs, USA alignment of a lidar for autonomous driving every vehicle during SOP and later on at random. To the autonomous car is almost here. Autonomous in urban scenarios is currently the ultimate of unprecedented scale. Mike Tzamaloukas, vice Enabling automated driving in inclement weather Richard Sands, application engineer, include a maximum number of sensors and actors, mobility will exist in many shapes, and insurance challenge. Statistical approaches are neither president, BU Autonomous Drive, ADAS and is challenging because sensor performance is Oxford Technical Solutions Ltd, UK autonomous parking procedures are chosen as test companies need to prepare for self-driving and feasible nor comprehensive. The talk will Navigation, Harman will present a novel framework compromised. For example, a light snowfall that This presentation covers a recent case study that scenarios. The action of the vehicle is observed by the new risk factors. During the transition, the address current views of challenges and that ensures that every connected car on the road has covers lane markings would render most methods addresses the challenges faced determining precise a micro GPS system. The quality is automatically transfer of responsibility must be designed with approaches in releasing self-driving vehicles. the ability to benefit from validated, crowdsourced of lane keeping inoperable. However, there are new alignment of lidars in an inertial navigation system identified by a complex fuzzy-logic-based algorithm. safety as a high priority. Insurance products and information. In this presentation, Mike will address methods for localisation that heighten the performance frame. We have developed a reliable calibration processes will need to be adjusted to account for More keynote speakers to follow – see topics spanning efficient multi-sensor data collection, of automated vehicles. In this presentation we will method to remove systematic errors caused by 09:30 - Road attribute confidence autonomous driving. Allianz has already gathered the website for the latest updates normalisation and machine learning cross-validation, share our experience using precision map data to mechanical misalignment. We will share techniques values using HD maps on-road experience by insuring autonomous spatio-temporal confidence estimation, data sharing improve the AV performance in poor weather. By using used to validate the calibration result, along with Michael Laur, senior technologist vehicles globally. Not every accident can be and customer privacy. He will also offer insights on how different methods of localisation against the precision specific findings that will help to improve confidence automated driving, Delphi, USA prevented by an ADAS, and a standardised data this technology can help reduce production quality map, VSI will discuss how it uses HD maps to improve in open-road sensor testing and validation projects. The presentation will discuss a method storage system in automated vehicles is required. costs and accelerate go-to-market timelines for OEMs, the performance and safety of automated vehicles, of adding probabilistic confidence values fleet operators and transportation city planners. even when lane markings are covered or absent. 17:30 - Challenges of multi-sensor and fusion ECU measurement for ADAS Level 3-5 14:45 - NDS and OADF map standards 15:45 - 16:30 - Break Alfred Kless, business development manager, 14:15 - 18:00 - Room A - Test, Verification & Validation for successful AV deployment Vector Informatik, Germany Dr Volker Sasse, VP Navinfo / chairman NDS 16:30 - Integration of solid-state The presentation shows the challenges and solutions and OADF, Navinfo & NDS, Germany lidar in vehicles: best practices for a complete ADAS logging approach: multichannel 14:15 - Map-based validation showing advantages and limitations identified vehicle test infrastructure, and track testing provides This presentation will address NDS and OADF map for superior object detection vehicle bus logging – CAN-FD, FlexRay, LIN, of autonomous vehicles in selected demonstrators of the project. real-world sensor data required to thoroughly standards, which are key for successful deployment Filip Geuens, CEO, XenomatiX, Belgium Auto-Ethernet; radar measurement technology for Dr Henning Lategahn, CEO, exercise the entire active safety system from sensor of autonomous vehicles. History shows that the map Technology choices for automotive lidar must be based raw data as well as object data; high-end fusion Atlatec GmbH, Germany 15:15 - Autonomous drive controllers to software. However, system-level testing often was always of high importance. However, AVs require on how and where the lidar units can be integrated in ECU measurement with multi uC + uP architecture Autonomous driving will hit the road in the near challenge validation and verification requires sensor fidelity that is difficult to capture in the support of better maps. Therefore, the whole vehicles. This integration impacts the lidar technology based on Autosar and Autosar adaptive operating future. One of the obstacles that is still to be test data management simulation, and extensive track testing is incredibly world will need be mapped extremely precisely. For as well as the vehicle itself. As the need for reliable solid systems; multiple vehicle and reference camera overcome is the validation of this technology. Alexander Noack, head of automotive expensive. By combining real-world logging with lab- example; Japan already collected all freeway maps for state lidar systems continues to grow, considerations measurement; other sensors such as laser scanner, How can one prove that the vehicle senses the electronics, b-plus GmbH, Germany based HIL simulation, you can generate the highest- autonomous driving at Olympia 2020. But AVs require about sensor positioning and ways to achieve reliable ultrasonic and GPS. The unique solution covers environment correctly? Two approaches are Centralised computing platforms for autonomous fidelity data streams and play them back in a variety worldwide standard maps and more cooperation detection are gaining automotive attention. This two use cases: ‘Engineer Mode’ with full graphical currently followed. One is real-world driving tests driving functions – the domain controllers – require of controlled scenarios to maximise test coverage. because each car acts on map data collection; other presentation will report on some of the outcomes from object overlay display, and ‘Taxi Driver mode’. and the other is virtual testing in simulators. Both a huge amount of input data. Multiple cognitive updates are too slow. Up and down streams will fill the the research cooperation with Tier 1 partners for specific seem tempting. The first, however, suffers from sensor sources in autonomous cars challenge 17:00 - Unified system of tools for ADAS the unavailability of accurate ground truth data, the whole testing and validation scenario. Huge Dr Florian Baumann, technical director, whereas the latter suffers from unrealistically computing power processes gigabit data that Adasens Automotive GmbH, Germany 14:15 - 18:30 - Room C - Testing Autonomous Vehicles in Real-Life Environments simplified virtual worlds. We present an approach forms the basis for decisions. Conventional ECU We are proposing a unified system of tools to to acquire accurate 3D map data that can be used development methods such as prototyping efficiently test and validate computer vision as ground truth and realistic 3D world models. platforms, vehicle loggers and simulation systems algorithms. This includes the categorisation of 14:15 - Complete approach for testing 14:45 - Using the StreetWise scenario 15:15 - London’s Smart Mobility Living Lab such as hardware in the loop are reaching different self-developed tools for pre-development, automated vehicles on a testing ground base for virtual safety assessment Iwan Parry, head of connected 14:45 - Experiences gathered technological limits. The presentation focuses development and post-development stages. The Dr Houssem Abdellatif, global head autonomous Sytze Kalisvaart, project manager integrated and autonomous vehicles, TRL, UK from the application of the on the changes in sensor architecture and the pre-development stage consists of recording and driving and ADAS, TÜV SÜD, Germany vehicle safety, TNO, Netherlands This presentation will provide a review of London’s ENABLE-S3 V&V architecture challenges of methodology for measurement data planning tools. The development stage consists of The physical testing of highly automated vehicles For virtual safety assessment of automated driving Smart Mobility Living Lab, part of the UK Government’s Dr Andrea Leitner, research project manager management in the domain controller environment. tools to visualise output of cameras, lidar or radar. is a very challenging task. Highly accurate and vehicles, a set of test scenarios is needed with real- £100m investment in creating an integrated CAV ADAS, AVL List GmbH, Austria The post-development stage is to test KPIs such repeatable replay of complex manoeuvres must be world validity. The StreetWise scenario database testbed ecosystem for on- and off-road testing. The This presentation aims to give an overview of 15:45 - 16:30 - Break as ROC curves, false positives, etc. This talk will achieved on the test ground to judge the ability of is based on real-world driving data. Parameterised Smart Mobility Living Lab will build on established first results of ENABLE-S3, a large European inspire people to develop such tools internally. autonomous vehicles to cope with different situations. observed variants of the scenarios are stored. and ongoing CAV projects in London, including project with around 70 partners from industry 16:30 - Maximising test coverage All tools are running via a web front-end, and For this purpose, we present a complete technical When using the database for virtual testing, the GATEway and MOVE_UK, and provide infrastructure and research. The project’s goal is the provision of utilising high-fidelity vehicle record demos will be shown during the presentation. setup that has been developed by TÜV SÜD and test engineer will have to select a relevant set of to support real-world testing, development and technology bricks (methods, tools and models) and and playback techniques its partners. Soft static targets, remotely controlled scenarios for the system under test. A data-driven evaluation of CAVs and CAV mobility solutions together a generic test architecture to enable more efficient Dragan Jadric, principal product manager, 17:30 - Systems approach to traffic simulation vehicles, a vehicle control system, approach for determining the relevance of the with new ITS and communications technologies. testing of automated systems. Standardisation National Instruments, USA creating ‘interesting’ test scenarios highly sophisticated sensors and a monitoring scenario is proposed. Also, the interface ‘scenario is a major enabler for modularisation and As autonomous functionality in vehicles becomes for autonomous vehicles system work together precisely. Technical details database – simulator tool’ and relationship to 15:45 - 16:30 - Break reuse. Therefore, the consortium evaluates more sophisticated, the test coverage required Siddartha Khastgir, PhD researcher, WMG, and demonstrations are provided in this talk. key performance indicators is presented. and extends existing specifications such as continues to expand. Traditional techniques such as University of Warwick, UK OpenScenario and OpenSimulationInterface. hardware-in-the-loop testing provide an excellent To prove that automated driving systems are safer The talk will highlight some first results foundation upon which to establish an autonomous than human drivers, it is suggested that they will

2 www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 3 Autonomous Vehicle Test & Development Symposium

10% EARLY-BIRD DISCOUNT DAY 1 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 2

16:30 - Importance of real-life deployment blossoming worldwide, and are crucial to address municipalities face high costs of bus lines. These of autonomous vehicles in mobility services challenges inherent in public acceptance and the seem inefficient because buses are used intensively 12:15 Panel Discussion – Harnessing the power of simulation Raphael Gindrat, CEO, BestMile, integration of autonomous vehicles in urban areas. during rush-hours but at other times by only a handful Switzerland The presentation will offer a deep dive into existing of people. Nonetheless, municipalities need to offer Simulation is fundamental to validating autonomous driving application. In this discussion, we will reach further into virtual environment, SIL/HIL and virtual data creation. The democratisation of autonomous vehicles will autonomous vehicle deployments around the world. basic transport facilities. Because the main expense be directly impacted by the quality of the resulting is labour costs, some municipalities are considering Robin van der Made, product manager Jon Horsley, Caius Cioran, mobility services and their integration in the current 17:00 - Exploring automated bus automated buses within their towns. To optimise software and services, TASS International programme director, application engineer, transportation infrastructure. In June 2016, Switzerland lines and cooperative driving availability, costs and reliable operation, a system (a Siemens Company), Netherlands Digital Engineering & Test Centre, UK dSPACE GmbH, Germany was one of the first countries to set up a service of Dr Arie P van den Beukel, assistant professor, design is being explored with small bus units that electric autonomous shuttles circulating through a University of Twente, Netherlands individually drive autonomously in dedicated lanes and city centre in pedestrian areas and on open roads. Reliable automation within complex traffic situations, drive cooperatively with a human lead driver in town. Today, deployments of autonomous shuttles are especially in cities, is challenging. For city transport, 17:30 Panel Discussion – Get the maximum from real-world testing 14:15 - 18:00 - Room A - Robust Test, Verification & Validation Methodology

High-quality data generation, public perception, repeatable real-world scenarios – we discuss how to get the most from your real-world testing programmes. 14:15 - Processes, tools and philosophy to presentation proposes a conceptual solution to 17:00 - Applying proven methods for Dr Houssem Abdellatif, global head autonomous Sytze Kalisvaart, project manager integrated Raphael Gindrat, CEO, develop an autonomous electric truck these problems of formalisation and evaluation. quantifying test results and test coverage driving and ADAS, TÜV SÜD, Germany vehicle safety, TNO, Netherlands BestMile, Switzerland Joachim Fritzson, CEO, Zuragon, Sweden Rainer Straschill, ADAS/AD strategist, FEV As a new way of thinking about road transport, Einride 15:15 - Challenges for testing with platform Europe GmbH, Germany was facing a lot of challenges. New powertrain robots at high speeds (+100km/h) The presentation will show that questions of test technologies had to be merged with autonomous Markus Schmidl, COO, DSD Testing, Austria coverage for self-learning systems can be solved: any driving logic and new sensor technologies. This Test speed demands for active safety system self-learning system in our scope, including those required the freedom to choose processes and testing are continuously rising. Current requests with non-deterministic portions, can be identically tools without having to compromise with legacy from the automotive industry are up to 130km/h. transformed so that the aspects of self-learning and development, but at the same time facing the UFO currently reaches speeds of 100km/h as non-determinism can be considered identically to pain of starting to work from a blank sheet. This a first platform test. The presentation shows the problem of the unknown use case population. DAY 2 WEDNESDAY 6 JUNE presentation discusses the learning curve up to now, the challenges for the test equipment and the Next, the presentation will explain how established and how ViCANdo has helped in accelerating the proving ground of testing at these high speeds. and proven approaches from different domains process and balance risk, and cut development and (experimental physics, mathematical statistics, data validation time by being multi-platform capable. 15:45 - 16:30 - Break analysis) can be applied to achieve a measure for test coverage in the situation of populations where 09:00 - 12:45 - Room A - Validation in the Virtual Domain 14:45 - The structure of test specifications 16:30 - Verification techniques for safety and only basic knowledge about their properties exists. of highly automated driving functions security in autonomous vehicle software Dr Hardi Hungar, team leader verification Dr Mike Bartley, CEO, Test 17:30 - Testing the connected vehicle 09:00 - Training and validating into automotive testing for the first time to achieve driving have to be test-driven hundreds of millions of and validation methods, German Aerospace and Verification Solutions Ltd, UK Rosario Trapero, manager Connected Car automated driving applications using the required test quality at much reduced time and kilometres. This is possible using virtual test drives and Center (DLR), Germany The Innovate UK-funded CAPRI project brings together Competence Center, Dekra, Spain physics-based sensor simulation cost. This paper will outline this exciting coordinated SIL/HIL simulation only. In this context, appropriate Currently, scenarios are considered as main an experienced consortium of partners from industry, The connected vehicle landscape is complex, with Robin van der Made, product manager software activity and examine the digital London space. sensor models and integration options of sensor ECUs elements in test specifications of highly automated academia and local authorities, working together to different communication protocols and standards and services, TASS International (a Siemens with HIL test benches are the key. This presentation driving functions. A scenario captures a large deliver a complete end-to-end POD mobility service. involved. Successfully implementing and verifying Company), Netherlands 10:00 - Object detection and classification outlines a new sensor simulation concept and a number of concrete test cases in a formal way. The consortium aims to collate sufficient evidence from the variety of safety applications in the connected One of the latest needs in the area of automated based on virtually trained neural networks generic interface unit to insert raw data behind sensor It must permit these test cases to be generated the deployment trials and simulation testing to support car requires testing at several different levels, driving is the generation of sensor data as input for Ronnie Dessort, simulation consultant, front-ends. A unique radar sensor model is introduced. automatically. And the collection of scenarios must PODs becoming a recognised vehicle classification including the physical layer, the protocols and deep neural networks for the purpose of training TESIS DYNAware GmbH, Germany cover all behaviour that needs to be tested. The for use on public roads. This presentation will discuss: the algorithmic functionality of the applications. In automated driving applications. The PreScan In autonomously driven vehicles, object detection 11:45 - Traffic simulation for connected and scenario language needs sophisticated constructs to verifying and validating the safety and security of this paper, a comprehensive approach for testing simulation platform can be used to generate virtual and fusion of sensor information enables the vehicle autonomous vehicle technology testing cover the complex, reactive patterns of road traffic. the next generation of PODs for the on- and off-road all these different layers is presented, covering sensor data of all sensor technologies relevant to to perceive its environment and decide for certain Dr Alex Gerodimos, executive director, And the test specification must allow overlapping environments, and collating sufficient evidence to laboratory and in-the-field testing procedures. automated driving, such as camera, radar, lidar, driving manoeuvres. The virtual development of Aimsun, USA of cases to be systematically coped with. The support PODs as a new vehicle classification. ultrasound and DSRC. In this presentation we also such systems allows the comfortable definition and Connected and autonomous vehicles (CV/AV) have present the added value of injecting synthetic sensor reproducibility of a large number of traffic scenarios. In the potential to tackle challenging problems related data directly into platforms such as the Mentor this contribution, a publicly available state-of-the-art to safety and efficiency. Accidents, congestion 09:00 - 12:45 - Room B - Open-Road & Real-World Testing Graphics DRS360 and Nvidia Drive PX2 for virtual algorithm based on deep learning is trained and tested and pollution in urban areas could be substantially validation of automated driving applications by in a virtual 3D world. Special focus is put on variation reduced if these technologies were adopted on a means of hardware-in-the-loop (HiL) simulation. of different harsh environmental conditions such as large scale. One of the main challenges for transport 09:00 - Testing autonomous vehicles in 09:30 - Real-life testing under experimental ahead to enable faster and better administrative rain or soiled traffic signs. The results show that the agencies and technology developers alike is to harsh winter conditions – key findings clauses – recent regulatory developments processes to facilitate such ‘real laboratories’. 09:30 - Virtual twin of the presented development method can be used as an find ways to evaluate the performance and impact Harri Santamala, CEO, Sensible 4 Ltd, Finland Dr Alexander Duisberg, partner, London CAV testbed appropriate complement to common development. of such technologies before they are deployed. Sensible 4, a startup focusing on automation Bird & Bird LLP, Germany 10:00 - The UK’s CAV testing ecosystem Jon Horsley, programme director, Simulation provides a comprehensive, repeatable, technologies under harsh conditions and Sandbox testing under real-life conditions only works Michael Talbot, head of industrial strategy, CCAV Digital Engineering & Test Centre, UK 10:30 - 11:15 - Break cost-effective and near-limitlessly scalable testing varying environments, will take its self-driving if the regulatory framework strikes a good balance (UK Government’s Centre for Connected and The UK Government has recently funded a coordinated environment. As such, simulation provides an vehicle to northern Lapland for testing during between incentivising innovation and warranting the Autonomous Vehicles), UK set of physical CAV test facilities between London and 11:15 - New sensor simulation excellent digital complement to field testing for the winter of 2017-2018 as part of the Aurora safety of others. Federal government is aware of the CCAV presented an overview of the UK’s new Birmingham to provide CAV technology developers concept for virtual test drives regression and pre-deployment scenarios. Arctic Challenge project. Perception, remote challenges to enable quick and efficient administrative CAV testing ecosystem programme at the AV from across the world with the facilities to develop Caius Cioran, application engineer, control and multi-sensor redundancy and processes, which may touch on all sorts of regulations. Symposium in 2017. Since then we’ve announced and test the essential underlying technology and dSPACE GmbH, Germany 12:45 - 14:15 - Lunch solutions will be studied in open-road tests. The ‘Experimental clauses’ and related approval processes the winners of the first £51m competition (matched actual vehicles on public roads and safe off-highway The market introduction of autonomous vehicles is on presentation will describe field-test arrangements, are making their way into the framework. The by industry to £102m) and launched MERIDIAN, environments. Alongside the physical, we are the horizon. A crucial element of these systems is the technology, key findings and results obtained. presentation analyses examples around autonomous the ‘CAV hub’ that will coordinate the UK’s building the virtual testing environments and bringing reliable environmental perception by means of camera, drive and other areas of testing from a legal and capabilities and promote them internationally. innovative new technology from gaming, AI, etc. radar and lidar sensors. Functions for autonomous regulatory perspective, and how government is pushing

4 www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 5 Autonomous Vehicle Test & Development Symposium

10% EARLY-BIRD DISCOUNT DAY 2 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 2

10:30 - 11:15 - Break 11:45 - Controlled field testing of automated 12:15 - Regulatory testing of autonomous and connected driving in urban environments vehicles for trials in residential areas 09:00 - 18:00 - Room C - Connected Vehicle & Infrastructure Technology 11:15 - Real-world driving scenario Rico Auerswald, research fellow, Fraunhofer Niels de Boer, programme director, identification for AV functional safety Institute for Transportation and Infrastructure CETRAN, Nanyang Technological Gildas Thiolon, engineer data Systems IVI, Germany University, Singapore 09:00 - Intelligent traffic lights: NEW covered, and an extension is planned in 2018 with 11:15 - Aurora Borealis Intelligent Corridor scientist, Vedecom, France Current test and validation procedures for automated Singapore is embarking on a roadmap towards from testing to deployment urban-interurban use cases. The company is also for snowtonomous driving and CAD testing Knowledge of the real-world driving environment driving functions use field tests on public roads deployment of autonomous vehicles as a part of the Jeannet van Arum, director smart mobility, participating in the deployment phase of an automated Reija Viinanen, director Aurora collaboration, involving manually driven cars is fundamental, primarily to identify unknown critical scenarios. transport system in residential areas. To extend the Province Noord-Holland, Netherlands proving ground and test zone in Zalaegerszeg. Finnish Transport Agency, Finland allowing detailed identification of safety issues These scenarios are then validated in simulations trial of AVs from the existing testbed area to residential Harm Jan Mostert, senior advisor smart mobility, Aurora Borealis Intelligent Corridor is a digital cross- encountered by autonomous cars in similar complex or on test sites. The complex traffic environments areas, a ‘milestone 2’ test needs to be passed, after Province Noord-Holland, Netherlands 10:00 - A high-integrity, automotive- border corridor on E8 from Norway to Finland. The situations. The MOOVE project has been created to required to validate automated driving functions in which the licence to test AVs will be amended and The province of Noord-Holland is working on V2X grade antenna system optimised main objective is to enable C-ITS and CAD testing in study this problem. This presentation: explains the urban traffic may not be investigated with sufficient the permission to test in defined residential areas will communication. By putting in place intelligent traffic for V2X DSRC communications arctic conditions. The road has been instrumented context and the developed methodology applied detail outside of field tests. Therefore, we present an be added. This presentation will give an overview of lights, which are able to communicate either through Dr Oliver Leisten, technical director, with supporting infrastructure to promote car to the collected data; presents the data modelling approach that increases the reliability of field tests by the test requirements, the test definition and technical wi-fi P or 3G/4G, use cases can be carried out. Industry Helix Technologies Ltd, UK makers’ CAD trials on public roads. E8 provides an applied to build the database; identifies the role of generating scenarios considering real urban roads assessment methodologies, and the work to improve and government identified ‘Day One’ C-ITS use cases The safety-critical role of V2X DSRC in driver-assist opportunity to test cross-border ITS solutions and crucial variables interfering in the extracted real-world with public traffic, focused resource coordination for and extend the testing based on feedback from as the first promising ones for deployment. A couple and autonomous vehicle systems requires particular snowtonomous driving in arctic conditions, and driving scenarios. Thus, improved robustness will be the test procedure and cooperation with connection trials of autonomous vehicles in residential areas. of these Day One use cases relate directly to traffic attention to ‘real-world’ channel/propagation research what kinds of infrastructure and roads will be achieved for the autonomous vehicle architectures and infrastructure in the Digital Testbed Dresden. lights. In the last two years the province of Noord- challenges which to-date have not been successfully needed in the future in terms of automated vehicles. systems being developed at the Vedecom Institute. 12:45 - 14:15 - Lunch Holland (regional road authority in the Netherlands) addressed by incumbent antenna technologies. Helix combined pilots and deployments for these use cases. Technologies is developing an antenna based on 11:45 - Solutions for V2X test and This presentation will focus on the lessons learned. dielectric-loaded, multi-filar helix antenna technology validation for the connected car 14:15 - 18:00 - Room B - Using Simulation to Advance System Design & Validation that is optimised for use in the V2X DSRC IEEE 802.11p Axel Meinen, technical sales manager, 09:30 - Infrastructure for connected environment and employs the use of a diversity system S.E.A. Datentechnik GmbH, Germany and automated driving in Hungary architecture. The V2X DSRC antenna to be developed This presentation will focus on necessary tools for test 14:15 - Real challenges for simulation 15:15 - Key simulation features for 17:00 - Massively parallel simulation for Adam Nagy, traffic engineer, Hungarian Public by Helix Technologies will employ two ceramic-based, and validation of upcoming mobile communication in verification and validation of AVs autonomous vehicle validation testing and validating autonomous vehicles Road, Hungary compact, dielectric-loaded, multi-filar antenna elements like 802.11p or cellularV2X (LTE-V) for connected Dr Roberto Ponticelli, chief engineer - Thomas Nguyen, R&D projects and AD/ADAS Christopher Hoyle, technical Future mobility requires more and more online co-located so as to provide optimum antenna diversity, autonomous driving vehicles. S.E.A. has developed intelligent mobility, Horiba MIRA, UK product manager, AV Simulation, France director, rFpro, UK information even from the road. The Hungarian promote link reliability and eliminate coverage nulls. a test platform to test and validate the technologies The ever-increasing use of virtual design, verification Recent projects in France and Germany show that Over the last two years we have developed a scalable, Public Road company is committed to supporting this and the corresponding hardware like ECUs and and validation (vDV&V) tools is supported by a simulation will play a key role in demonstrating fast, open simulation environment for autonomous development. C-ITS deployment started in 2015 in 10:30 - 11:15 - Break onboard units with respect to RF communication sound base of successful case studies across most the safety of autonomous vehicle systems. The vehicle testing and validation. Multiple ego vehicles Hungary to demonstrate the use of C-ITS to exchange and compliance, protocol analysis and validation, engineering and scientific areas (e.g. medicine, key features for these purposes are scenario and may share the same simulated real-world environment, data through wireless communication technologies as well as complex scenario generation including aerospace and automotive). Nevertheless, there sensor models. At AV Simulation, a new joint venture each with multiple sensor model feeds. The entire between vehicles and infrastructure (V2I). Road safety GNSS simulation for open-loop or closed-loop exists a risk of a false sense of safety when the between Oktal and Renault, we have developed environment may run offline, for automated regression improvement, especially work zone safety, was the HIL tests. A brief overview of the different test tools used on vDV&V of safety-related systems a versatile simulation engine dedicated to virtual testing, or may run real-time, which also allows key issue, but the upgrading of services is ongoing. aspects will be given, and a way to analyse and are not well understood, particularly in highly evaluation and validation of automotive systems. multiple human drivers to join the simulation, from A 136km-long stretch of motorway M1 has been visualise the V2X test data will be demonstrated. complex systems like autonomous vehicles. This In this presentation, we will present key integrated driving simulators, adding stochastic, unpredictable, talk will address how the modelling methods, features: a powerful generation engine that enables error-prone behaviour, i.e. real human behaviour, into physical data and correlation, design and test a complete representative set of scenarios, based the simulation. The same simulation environment may 12:15 - Panel Discussion – The role of a transport authority in traceability and coverage, operational boundaries, on templates, and a complete tool chain for be shared by multiple OEMs or Tier 1s to evaluate how developing and implementing V2X technology dynamic environments, hardware in the loop (HIL), sensor simulation, including phenomenological, their autonomous vehicles behave when they interact. scenarios annotation and generation, and other key statistical and full-physical modelling. parameters shape the challenge for AV vDV&V. 17:30 - Virtual testing by coupling system Many believe that true L5 AVs cannot be achieved without C-ITS. Here we will discuss the role of transport authorities in creating a robust V2X network. 15:45 - 16:30 - Break simulation with SUMO traffic flow simulation 14:45 - Enabling the massive simulation Dr Jakob Kaths, product manager, Jeannet van Arum, director Harm Jan Mostert, senior advisor Adam Nagy, traffic engineer, Reija Viinanen, director that autonomous driving validation requires 16:30 - Purpose-driven scenario generation TESIS DYNAware GmbH, Germany smart mobility, Province smart mobility, Province Noord- Hungarian Public Road, Hungary Aurora collaboration, Finnish Enguerrand Prioux, ADAS/AV product line Dr Andreas Höfer, product manager simulation Increasing automation in the automotive sector will Noord-Holland, Netherlands Holland, Netherlands Transport Agency, Finland manager, Siemens, France software, IPG Automotive GmbH, Germany lead to a stronger interaction of individual vehicles Proper SAE Level 4-5 vehicle validation will ADAS have undergone a rapid evolution. Formerly with surrounding motorised and non-motorised require – at the least – millions of scenarios to be integrated as standalone components, modern ADAS traffic and with traffic infrastructure. This creates new checked. For project timing and testing conditions are networked with each other, which causes new requirements for virtual development and testing reasons, this will be entirely feasible only through challenges for development and testing. This calls that can be met by integrating automotive and traffic simulation. To validate sensor technology and design, for new testing techniques, such as scenario-based engineering simulation tools. To this aim, an approach computer vision and sensor fusion, the decisional simulation. There are three basic ways to create is presented to seamlessly couple high-fidelity models 12:45 - 14:15 - Lunch transit signal priority (TSP) solution. This centralised test CAV technologies for traffic intersections. As and executive driving agent, and actuation, most scenarios: outlining on the drawing board and manual based on Matlab Simulink and the microscopic solution demonstrates how cities and companies can part of the project, a smartphone app is developed companies will have to couple several simulation creation with a scenario editor; generation based on traffic flow simulation SUMO. This is achieved by 14:15 - Leveraging V2I communications use connected hardware to create a reliable network for integrating cyclists and pedestrians (vulnerable models (AI, controls, world, sensors, vehicle map data with optional enhancement in a scenario establishing a co-simulation setup between the two to enable fully autonomous operations that is capable of supporting autonomous vehicles. road users or VRUs) into the overall cooperative CAV physics) and perform a massive simulation activity editor; generation based on real-world measurements tools while using common map information based Mike Haldane, vice president of global system. This paper presents the existing technological to sufficiently sample the scenarios space. In this with optional enhancement in a scenario editor. Each on the OpenDRIVE road description format and marketing, Global Traffic Technologies, USA 14:45 - Supporting vulnerable road users in approaches for integrating VRUs in cooperative presentation we will communicate Siemens’ progress approach has its own strengths and weaknesses, including driving dynamics, controllers and in-vehicle For nearly 50 years, Global Traffic Technologies’ connected and automated driving landscapes CAV systems, our adopted approach as well as in this area through a concrete simulation case. which are described in this presentation. sensors such as camera, radar, lidar and ultrasonic. traffic signal priority control technology has been Dr Fatih Özel, project manager and future development and implementation steps. connecting vehicles to infrastructure. This technology consultant, Oecon Products and NEW has evolved from IR to GPS-enabled radio and cellular- Services GmbH, Germany 15:15 - Active pedestrian safety based communications. The V2I communication Connected and automated vehicles (CAVs) are the testing in CARISSMA infrastructure that enables driver-assisted priority subject of extensive research nowadays owing Dr Igor Doric, senior researcher BOOK YOUR PASS ONLINE NOW! 10% EARLY-BIRD DISCOUNT control can be extrapolated to autonomous vehicles to their potential to improve safety. The Federal and deputy scientific and technical as technologies advance. GTT’s latest application Ministry of Transport and Digital Infrastructure in manager, CARISSMA, Germany 3-DAY PASS €1,950 €1,755 + VAT 2-DAY PASS €1,750 €1,575 + VAT in New York City leverages existing connectivity on Germany has funded the Digitaler Knoten 4.0 (Digital The presentation will discuss the results of the buses and at intersections to create a centralised Intersections) project, which aims to develop and pedestrian safety research projects TARGETS

6 www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 7 Autonomous Vehicle Test & Development Symposium

10% EARLY-BIRD DISCOUNT DAY 2 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 3 and VISYTE. The projects ran from October 2013 16:30 - Pedestrian in the Loop related systems are not adequately tested, e.g. 11:15 - 16:15 - Room A - Using Simulation for Test Scenario Validation to October 2015, and from November 2015 to Michael Hartmann, research engineer, in collision avoidance scenarios with pedestrians. January 2018 respectively, and were supported Virtual Vehicle Research Center, Austria Examples are the change of pedestrian behaviour by the Federal Ministry of Economic Affairs and A large number of testing procedures have been caused by interaction, environmental influences 11:15 - Digital simulation is key to approval procedure. To this end, a safety assessment situations. The new high-performance ROADS Energy, on the basis of a decision by the German developed to ensure vehicle safety in common and personal aspects, which cannot be tested in real autonomous driving safety framework that employs virtual assessment of traffic dynamic driving simulator, commissioned to AVS, will Bundestag. Furthermore, the presentation will provide and extreme driving situations. However, these environments. It is proposed to use a test environment Jean-Paul Roux, senior vice president, scenarios is proposed. This framework consists of help Renault test these cases in the most efficient details about the CARISSMA research and test conventional testing procedures are insufficient for with flying drones. Name: Pedestrian in the Loop. EU operations, Exa Corporation, UK four components: data acquisition, scenario extraction and realistic way and design the best experience centre, including the CARISSMA rain simulation. testing autonomous vehicles. They have to handle Digital anti-soiling simulation work is particularly and parameterisation, virtual safety validation and for the final client as well as providing help for unexpected scenarios with the same risk a human significant in the upcoming world of sensor-controlled physical safety validation. Due to the simulation- general driver acceptance and public approval. 15:45 - 16:30 - Break driver would take – or less risk. Currently, safety- fully autonomous vehicles. We explain how orientated nature of this framework, quantitative and PowerFLOW software incorporates dirt, dust and statistically relevant measures for safety-related AV 15:15 - Decision-making water into simulation – with the real-world accuracy performance are obtained while minimising the number evaluation by oracle vision 17:00 - Panel Discussion – How do we develop, test and verify AVs to protect vulnerable road users? that can only be achieved using a transient solver. of physical tests, thus realising an efficient procedure. Dr Rémi Régnier, researcher in evaluation of data We explain how the particle-handling capability The assessment framework is currently being processing systems, LNE, France Vulnerable road users and pedestrians create unique challenges for AV test programmes. In this discussion, of the software enables automotive and truck developed in Singapore to accelerate AV deployment. The SVA Project aims to meet the challenge posed we will evaluate current and future testing methods for VRUs and pedestrians. manufacturers to deliver a safer and much-improved by the complexity of demonstrating the safety of driving experience through better control of soil and 12:45 - 14:15 - Lunch autonomous vehicles through the use of digital Dr Fatih Özel, project manager and consultant, Dr Igor Doric, senior researcher and deputy scientific Michael Hartmann, research engineer, Virtual water accumulation on autonomous driving sensors. NEW simulation. One of the objectives is to provide builders Oecon Products and Services GmbH, Germany and technical manager, CARISSMA, Germany Vehicle Research Center, Austria 14:15 - Realistic human behaviour in and suppliers with a methodology and simulation 11:45 - Dynamic ground truth measurement simulation: the problem no-one has cracked tools for the validation of safe autonomous vehicles. for simulation and post-processing Kirsty Lloyd-Jukes, managing We propose a new methodology based on the Steffen Metzner, technology scout director, Morpheus Labs, UK concept of oracle, a mathematical tool adapted to the ADAS, simulation and control, We all know simulations will be critical for AV testing. ADASS evaluation. Indeed, oracle created an ideal AVL List GmbH, Austria But there’s one problem that no-one’s cracked yet: reference to compare with the answer of a system. The For development of ADAS and AD functionalities, use how to accurately represent human behaviour (driver, presentation will explain how to create this oracle. of environment simulation tools is a well-accepted cyclist, pedestrian) in a simulator. Traditional methods DAY 3 THURSDAY 7 JUNE methodology. Most expensive and time consuming like swarm don’t provide sufficient fidelity to really 15:45 - Virtual automotive testbeds is the manual definition and design of complex put an autonomous car through its paces. After all, if for automated, reproducible tests scenarios as input to the experiments. The DGT humans always behaved in a logical and rule-abiding Jörn Thieling, research assistant, Institute (Dynamic Ground Truth) research project investigates way, human error wouldn’t be responsible for 95% of for Man-Machine Interaction at RWTH methodologies and measurement systems for the accidents. This presentation explores an alternative Aachen University, Germany 09:00 - 10:30 - Room A - Using a Test Facility to Advance Testing Programmes capturing of dynamic objects surrounding moving approach: how a state-of-the-art machine learning Today, autonomous vehicles are mainly verified test vehicles without the need for invasive changes technology called ‘learning from demonstration’, in real-world test drives. These tests are not only (e.g. to mount sensors). The recorded data should normally deployed in the field of social robotics, time-consuming and expensive but also insufficient 09:00 - Introducing the UK’s 09:30 - Proving Ground Zala – a unique 10:00 - Autonomous highway driving be usable for onboard processing (enabling direct can be applied to cracking this exciting problem. for a trustworthy validation, since critical situations controlled urban testbed for test environment for future mobility in undetermined weather conditions evaluation and comparison with the environment rarely occur. To enable efficient system tests, we connected and autonomous Zoltán Hamar, technical director, Automotive Dr Sergey Shadrin, scientific model of the vehicle) and offline post-processing 14:45 - High-performance driver-in-the- propose the use of virtual testbeds as a software Peter Stoker, chief engineer - vehicle, Proving Ground Zala Ltd, Hungary advisor, BaseTrack, Russia for more accurate and detailed data analysis. loop simulators for autonomous vehicles environment in which ‘digital twins’ of real vehicles, Millbrook Proving Ground Ltd, UK The objective of the project is to establish an Technology provides autonomous driving within Dr Andras Kemeny, expert leader, sensors and environments are simulated (from The critical need for a semi-controlled but realistic automotive testing site for the automotive and a road lane without visual recognition of the road 12:15 - Scenario-based performance Simulation & VR, Renault, France RWTH). That simulated data is compared with urban test environment for CAVs that seamlessly communication industry, which would be built on marking, while reaching speeds up to 130km/h assessment framework for For safe autonomous vehicle products for all, real-world data by using established algorithms connects with open-road urban environments, raises the domestic R&D capacity of the sector and on in undetermined weather conditions (rain, snow, automated vehicles validation on billions of kilometres is needed, thus for environment interpretation from the automotive unique challenges. Millbrook, in collaboration with the European R&D capacities equally. In line with current etc.), which is achieved by receiving high-precision Erwin de Gelder, scientist, TNO, using both massive and driver-in-the-loop simulation, sector (with ADASENS input and evaluation). United Kingdom Atomic Energy Authority’s RACE, and future automotive trends, when establishing navigational data. The composition of the technical Netherlands with a large number of driving scenarios. One of is working to address these challenges. Based on a the testing site we place special emphasis on the solution includes in-vehicle automation kit, hybrid In view of recent developments in autonomous the main critical scenarios is the handover between step change enhancement to existing roads and test testing needs of autonomous vehicle systems and navigation system, safety subsystem of risk vehicles (AVs), the need arises for an efficient AV road- manual and autonomous mode in different traffic capability, the facility will revolutionise preparation and related environmental conditions. Proving Ground assessment. The proposed approach provides highway validation of CAVs, services and technologies for public Zala is a unique test site where the fusion of classic autonomous driving and supports lane-keeping road deployment. It will offer open access to users – dynamic test elements and test elements of future assistance systems in cases of road marking visual 09:00 - 16:15 - Room B - Test Scenario, Verification and Validation Studies including developers of software, sensors, roadside technologies are realised on a 260ha area. recognition failures. Practical issues will be presented. units, telecommunications (5G) and cybersecurity systems – to explore public and industry impacts and 10:30 - 11:15 - Break 09:00 - ACU internal fault detection 09:30 - Secure testing of ADAS functions 10:00 - It’s all corner cases: teaching accelerate the development of CAV technologies. logic evaluation method using using a VRX driving simulator computers to drive safely semiconductor modelling Günther Hasna, director strategic projects, Bruno Fernandez Ruiz, co-founder Byoungmoo Kwon, research engineer, Optis GmbH, Germany and CTO, Nexar, Israel Hyundai Autron Co Ltd, Korea Optis is developing a simulator software for dynamic As the industry is charging towards self-driving “I have attended the Autonomous Vehicle Test & Development Symposium on The objective of our study is to present an evaluation ADB headlamp functions, named VRX-Headlamp. perception in all-terrain, all-weather and all-lighting, method through semiconductor modelling of peripheral Due to physics-based simulation algorithms it is it remains a challenge to collect and tag enough data multiple occasions and it never disappoints. The speakers are always well versed sensors, which efficiently evaluates the ACU’s internal possible to replace real night test drives with virtual to allow for reliable driving algorithms. Applying user in the latest industry trends and best practices, and being able to network with fault detection software logic. This study introduces simulations. As the ADAS function of the light-assist self-annotation of driving data is the only way we two different methods to generate internal fault system is dependent on sensor inputs controlling the can truly advance the autonomous car industry and others in the industry who are developing autonomous vehicle technologies is conditions, one by editing a pre-recorded data stream headlamp, it was also necessary to develop virtual infrastructure. In this session, I will share our concept incredibly beneficial. This event has its finger on the pulse of the autonomous vehicle of normal operation, and another by modelling the physics-based sensors as cameras and lidar. For safety and methods in creating an autonomous learning fault-occurring sectors and adding fault logic blocks ADAS functions like the autonomous emergency ‘flywheel’ of collect-annotate-learn road data network. development community, and it provides me with insight that I can use every day.” to the sectors. We evaluated a test MCU’s software braking AEB, it is now possible to do a virtual validation with 386 cases of 95 possible internal fault types of mandatory NCAP ratings like the pedestrian 10:30 - 11:15 - Break Nicholas Keel, group manager – DAQ and control product management, National Instruments and achieved a test time reduction of 70% compared AEB night tests with the VRX-Sensors software. with the current method of internal fault testing.

8 www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 9 Autonomous Vehicle Test & Development Symposium

10% EARLY-BIRD DISCOUNT DAY 3 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 3

11:15 - A case study of Level 4 NEW mixture of tests in real and simulated environments. purpose is to generate test cases automatically trials around the world as well as our own analysis 11:15 - Testing V2X systems by conducting the net. The future of autonomous driving relies test and validation in Korea Even after a functional decomposition in several from a test specification. Now test designers are to demonstrate reasonable metrics that can be used simulations and field tests simultaneously heavily on the availability of GNSS signals, both for Younggi Song, CEO, SpringCloud Inc, Korea classes of base scenarios there always exist many able to handle 11 times more test cases, deliver test to evaluate AV risks and human trust in the AVs. Frank Brennecke, managing director, Oecon navigation purposes and for sensor synchronisation, This presentation will cover self-driving cars and parameters with a huge number of possible values. results earlier and cover more requirements. Products and Services GmbH, Germany as well as for V2X synchronisation. It is essential to their test and validation procedure. The system Testing all combinations of values is impossible. In 10:00 - V2X communication – a C-ITS systems using V2X communications provide design autonomous vehicles for operation in jamming/ consists of multiple sensors and a controller for the presentation we discuss possible approaches and 15:15 - ADAS testing advanced: test tool’s point of view opportunities for achieving sustainable transportation. spoofing environments as well as including such target vehicles (conventional vehicle and EV). Based challenges in exploring the large parameter spaces 6D target mover Thomas Löffler, manager, Vector However, before any V2X system is fully developed facilities in the testing range. GPS Dome provides on test scenario L4, each task consisted of a ‘start, in the search for hidden faults and critical situations. Dierk Arp, executive director, Messring Informatik GmbH, Germany and deployed, an extensive phase of testing is commercial solutions for jammers/spoofers. driving, stop and parking’ procedure within the Systembau MSG GmbH, Germany V2X communication is a key enabler for autonomous required. Using dedicated private test tracks and test ground. The service point is from ground IoT 12:45 - 14:15 - Lunch Pedestrians and cyclists account for a significant driving and helps extend the functionality of advanced simulations are two main approaches for testing 12:15 - Autonomous drive and connected sensors and vehicle movement information and will proportion of road deaths worldwide. Current ADAS driver assistance systems (ADAS). The presentation V2X systems. In this paper, a method is proposed cars: a VTOX business case provide task performance as defined. The data from 14:15 - Assurance of autonomous vehicles test systems are tackling this challenge, but are limited introduces the limitations of today’s ADAS systems, for testing V2X systems by using both approaches Stéphane Barbier, chief development SDC is 5 type of dataset and will be continuously with authentic data recordings in their design to linear or two-dimensional motion. and shows why V2X communication is such a hot concurrently and utilising systems such as stereo officer, Transpolis, France monitored by platform service architecture. Bernhard Kockoth, technology scout, With this setup, particularly during acceleration topic and how it can support ADAS and autonomous cameras and roadside stations. The method is aimed The speaker will present a Transpolis business case ViGEM GmbH, Germany processes, an unrealistic motion is generated. The driving. It also discusses the measurement and to be implemented on a motorway section of the A39 with an OEM and a traffic-light manufacturer. GLOSA = 11:45 - Autonomous vehicle mode-based Autonomy driving Levels 3-5 are based on a growing concept of hanging dummies from above creates testing challenges involved inv coming up with ADAS near Brunswick, which is part of the Lower Saxony knowing the position and the current and future status testing: generate test cases for 3D simulator number of safety-related systems that must be new possibilities for more life-like dummy trajectories applications supported by V2X communication. Test Field for connected and automated driving. of the next traffic lights to enable the definition of an Fabrice Trollet, MaTeLo product secured with millions of kilometres. All vehicle bus using six degrees of freedom. The system sets new optimal eco-driving strategy. The driver is advised manager, All4tec, France communications and raw data from sensors, cameras, standards in precision and repeatability through 10:30 - 11:15 - Break 11:45 - GNSS jamming and spoofing with an optimal speed in order to get a ‘green wave’ Based on a Renault-Nissan methodology and other lidar and radar, as well as status data like weather the ability to reproduce real-life human motion Moshe Kaplan, CTO, GPS Dome Ltd, Israel (pass the traffic light with a green light state), or with a R&D projects, this presentation explains how to deal and actual maps, must be recorded, authentically. sequences and imitate them realistically – for example, During the last few years, the phenomena of slow-down and stop instruction. The objectives are to with the huge combinatorics met in ADAS simulation An 8hr test drive easily produces 4TBytes of data, if based on data from a motion capture system. jamming and spoofing the GNSS signals has become reduce stop times, reduce unnecessary acceleration and how to compute related autonomous vehicle not 20 to 100Tbytes. The data then must be fed to widespread. The source of jamming and spoofing in urban traffic, save fuel and reduce emissions. expected behaviour, to generate tests cases. The data centres without causing long pauses in vehicle 15:45 - Autonomous vehicle fleet testing may be on a national level and on a personal level. speaker will present the MaTeLo Model-Based Testing testing. The enormous amount of data becomes a – or how to deal with really big data GNSS jammers/spoofers are easily obtainable over 12:45 - 14:15 - Lunch process, in which weather, road infrastructure, other challenge for measurement equipment in automotive Dr Sebastian Bode, team manager, IAV motorist behaviour, EGO system, road events, etc. environments. New concepts and solutions are GmbH, Germany are designed. He will also explain how to design presented to meet sophisticated requirements. Leveraging the potential of fleet testing campaigns 14:15 - 16:15 - Room C - Best Practices determinist test cases with great variability. From this of autonomous vehicles equipped with multiple MaTeLo model, the speaker will show the automatic 14:45 - Meeting the high demand for sophisticated sensory systems such as lidar as well generation of thousands of scenery samples using automated tests – the TestCase Generator as communication systems connecting the vehicle 14:15 - Cognitive simulation of the driver exotic metrology technique for almost half a century, create better assistance systems, ultimately Gibbs sampler algorithms, to automatically build Tobias Weimer, specialist for system and software with the IoT, challenges even the most bleeding-edge for autonomous driving system tests until recent development in laser components, resulting in an increase in road transport safety. Oktal SCANeR scripts for 3D simulation run. development, MicroNova AG, Germany big data technologies. Getting the most valuable Jean-Charles Bornard, cognitive engineering and together with the intense interest in its autonomous The number of tests required for autonomous and insights into your data and optimising your engineering human factors, ESI Group, France vehicles applications, elevated lidar to the level 15:45 - Could we use DevOps methods 12:15 - Chasing critical situations electric vehicles has increased dramatically compared processes will require tools and services for data ESI is currently developing a cognitive driver model, of ‘enabling technology’. In this presentation, we to deliver autonomous cars? in large parameter spaces with established standards in the automotive industry. acquisition, management, analytics and visualisation. which will be integrated into its simulation solution illustrate the complementary capability between Guillaume Belloncle, smart NEW Dr Mugur Tatar, managing director, This demand can only be handled by means of This talk will discuss our testing along the way to Pro-SiVIC, in partnership with the French Laboratory radar and lidar while attempting to categorise the safe and connected car solution QTronic GmbH, Germany automation and adapting existing processes. Therefore, a solution that combines the best from the world of of Ergonomics and Cognitive Sciences applied to latest available lidar techniques by their wavelength, manager, Dassault Systèmes, France It is commonly agreed that testing and validating we developed a new solution for the test automation big data, cloud, AI and automotive engineering. Transport (IFSTTAR-LESCOT). Even though human measurement mode and imaging mechanisms. Model-based systems engineering (MBSE) methods highly automated driving functions will involve a software EXAM, called the TestCase Generator. Its behaviour and cognitive processes are still unsolved have been used for decades by the aerospace questions, human driver cognitive simulation will 15:15 - ADAS development on a industry and are increasingly being adopted within the allow a better understanding of the driver’s state driver-in-the-loop simulator automotive industry. However, they are sometimes 09:00 - 12:45 - Room C - V2X Testing & Validation and behaviour. Driver cognition simulation will give Jelle van Doornik, product manager ADAS, seen as too complex to continuously deliver disruptive autonomous vehicle engineers the ability to test their Cruden BV, Netherlands innovation for mass markets. Internet companies systems from main concept to HMI specification Recent research shows that many people are are continuously delivering innovations using agile 09:00 - Time-sensitive networking use cases of time-sensitive networking in vehicles and 09:30 - Managing risks of (human-machine interaction) or system validation, and not well informed about ADAS in their cars. This software development methods (a.k.a. DevOps), for autonomous driving also the ways to validate the network components autonomous vehicles these tests will integrate key performance indicators results in unnecessary and unsafe situations as but these methods are limited to delivering safety- Thomas Schulze, head of testing, automotive, such as ECUs and end devices like cameras and Kwang Sheun Tham, emerging technology of human-like performance and situation awareness. they learn to use the systems while driving, or do critical cyber-physical systems, like a self-driving Spirent Communications GmbH, Germany sensors, as well as the software applications lead, IAG Firemark Labs, Singapore not use the available systems at all. Companies car. We will outline how mixing these MBSE and Time-sensitive networking protocols and applications using this technology. Learn how autonomous Autonomous vehicles (AVs) have been pitched as 14:45 - A case study on accelerating focus on sensors and forget about the most DevOps views into an agile MBSE approach can are the enablers for using real-time functionalities in vehicles act like data centres on the road and how being able to significantly improve safety. However, lidar development and capability important aspect: the interaction between driver better support the virtual development and validation different industries based on Ethernet/IP networks. to avoid safety issues based on malfunctions. a mixed fleet scenario with AVs and conventional Mark McCord, vice president of and vehicle. Driver-in-the-loop simulators can of autonomous driving technology, leveraging For the automotive world, TSN will help to implement vehicles is deemed a reality for at least the next engineering, Cepton, USA be used to create a smoother shared control existing industry standardisation initiatives. driver assistance and autonomous functions in next- 10 to 15 years. This poses safety risks arising from Lidar was used for distance measurement shortly experience and better transitions when reclaiming generation vehicles. The presentation will illustrate interactions between these vehicles. We use data from after laser was invented in the 1960s. It remained an control. This enables automotive engineers to

BOOK YOUR PASS ONLINE NOW! 10% EARLY-BIRD DISCOUNT NETWORKING EVENING PARTY Wednesday 6 June, 18:00hrs, Exhibition Hall 5 3-DAY PASS €1,950 €1,755 + VAT 2-DAY PASS €1,750 €1,575 + VAT All delegates and speakers are invited to attend our ONE PASS – ACCESS ALL THREE EVENTS! complimentary networking evening. OVER 150 SPEAKERS PLUS OVER 80 EXHIBITORS

10 www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 11 10% EARLY-BIRD DISCOUNT DAY REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 1 DAY 1 TUESDAY 5 JUNE

09:00 - 12:45 - Keynote Opening Session

09:00 - How to get from expectations 10:00 - Benefits of augmented-reality 11:45 - Surf & Curve – HMI ready for SAE 3 to experience in autonomous driving head-up displays for automated driving Sören Lemcke, head of advanced human interface 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY Christoffer Kopp, product owner – UX autonomous Bettina Leuchtenberg, expert HMI, solutions, ZF TRW BCS, Germany drive, Volvo Car Corporation, Sweden ergonomics and usability, Continental NEW Within Surf & Curve, ZF-TRW has developed user- The presentation will discuss how we do early Automotive GmbH, Germany centred HMI concepts targeting automated driving of experience tests of autonomous driving to be able to Dr Thomas Vöhringer-Kuhnt, head of HMI SAE Level 3. Together with the strategic partners FKA INTERIOR DESIGN, DEVELOPMENT AND TECHNOLOGY guide the technology development in the right direction. user experience and design, Continental mbH and the Institute for Automotive Engineering It will focus on what elements are most important to Automotive GmbH, Germany of RWTH Aachen University, revolutionary HMI FOR AUTONOMOUS VEHICLES create trust in the autonomous car. It will also cover the Augmented-reality head-up displays (AR HUD) concepts were not only developed along several most important aspects to take into consideration to have been proved to support drivers during manual usability studies, but also tested and validated in the make the user feel freedom when in autonomous mode. driving. The presentation shows that drivers also high-fidelity and static driving simulator. Among other benefit from AR HUD content during autonomous features, Surf & Curve offers a unique way of lateral 09:30 - Shaping the future of driving situations. Based on an expert assessment, control with drivesticks, mirror replacement and comfort

NEW Autonomous Vehicle Interior Design & Technology Symposium UPDATED PROGRAMME 〉〉〉 automotive interiors – a vision relevant use cases and AR HUD elements have been control ready for automated driving of SAE Level 3. for innovative seating systems elaborated. In a driving simulator study with 24 users, Thomas Gunkel, design manager, industrial an interaction concept with AR HUD content was 12:15 - Creating ‘magic’ in the NEW design, Adient, Germany compared with an interaction concept without AR seats for autonomous driving Disruptive trends like autonomous driving, car sharing HUD content. The study shows that most users prefer Marc van Soolingen, Global design director or the growing urban population are drastically the AR HUD elements compared with information seating, Faurecia, France changing the means of transportation. Using a car in presentation in an instrument cluster. Takeover The presentation will cover: radical change – 2030 will be a completely different experience to what quality benefits from the use of AR elements. traditional premium and new premium; from mega we know today. Adient’s research and design teams trends to product directions; differentiating seating took these trends and other input data and translated 10:30 - 11:15 - Break value; values of today and values for tomorrow; them into the compelling AI18 seating concept, designing developments for new expectations; 10% which showcases various user scenarios and seating 11:15 - New challenges for HMI in transformer magic – beautiful structure, advanced solutions for vehicle transportation beyond 2030. the age of autonomous driving versatile structure, enabling new activities; new safety EARLY-BOOKING Rashmi Rao, senior director, advanced engineering, setup; digital magic – integrated screen, natural DISCOUNT! CoC user experience, Harman, Germany motion, fusion smart tiles; solutions for intuitive seat The age of the connected car has clearly demonstrated controls, technology in surface; integration magic Go online to book a need to raise the user experience to a new level. To – controlling complexity, seamless integration, your pass! ensure that autonomous driving is not one day equated home feeling in car, complete seat, UBA integration; with monotony, UX design is becoming the decisive health and wellbeing magic – short storytelling brand factor for OEMs. The presentation will outline around active wellness and closed-loop comfort; what a future-orientated HMI design including AR and conclusion –our development directions for AD cars. VR applications has to look like, and explain why the user experience (UX) will be one of the decisive factors 12:45 - 14:15 - Lunch in autonomous or semi-autonomous driving enabling a wide variety of interactions between man and machine.

14:15 - 18:00 - Afternoon Session

14:15 - The automotive cocoon 14:45 - Will parallel industries and 15:15 - Real-estate design Dr Dominique Massonié, product manager - non-traditional OEMs drive the opportunities of autonomous cars HMI, Elektrobit Automotive, Germany autonomous cars of the future? Richard Seale, lead automotive Impacted by rapidly evolving consumer electronics John Tighe, design director transport, JPA designer, Seymourpowell, UK technology and new automotive business models, the Design, UK The advent of autonomous technologies will have a REDEFINING car is now a connected object that is set to become an Through several years of experience and huge effect on the rule book that designers have to THE PASSENGER extension of someone’s home and office. This raises understanding of customer behaviour within aircraft play with, significantly throwing open the functional An integral part of an interesting question: what factors need to be taken and many other transport fields, JPA tells the story and aesthetic potential for our future vehicles and EXPERIENCE! into consideration when creating automotive interiors? of Person X’s life in 10-15 years; what their needs will travel. This shift will be largely led by both the change This presentation will provide insights into emerging be, what will be available to them, what is relevant in safety regulations, and the switch from a driver- Autonomous Vehicle challenges for automotive user interface design and and how this will influence the autonomous vehicle based model to a passenger-based model – bringing then focus on how in-car interfaces can be designed interior design of future. We will discuss the package unparalleled opportunities to customise our vehicles. TECHNOLOGY to deliver experiences that are specific to the user; the and seating layout, which is one of the most important The seismic shifts that autonomy will bring will WORLD EXPO 2018 tools one needs to create these interfaces; and the roles factors when designing successfully for autonomous open Pandora’s box, and nowhere will this be more usage-, user- and context-specific information will play. cars. What we’ve witnessed coming from automotive obvious than in the interiors of our future cars. We brands often seems to be blinkered and anchored in will explore original concepts in this presentation. PLUS HMI CHALLENGES AND NEXT-GENERATION the past, which brings risk of being leapfrogged by ONBOARD SERVICES AND TECHNOLOGY those outside the traditional automotive industry. 15:45 - 16:30 - Break

12 www.autonomousvehicleinteriors.com www.autonomousvehicleinteriors.com 13 10% EARLY-BIRD DISCOUNT DAY 2 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 3

16:30 - Mobile Livingroom 2.0 17:00 - Highly accurate reference system of a DMS system a new dimension of accuracy 14:15 - 18:00 - Afternoon Session Moritz von Grotthuss, CEO, Gestigon - a for validation of driver monitoring within the whole development process. Valeo brand, Germany Heiko Ruth, head of system department, Most experts expect the car to be the third place of CMORE Automotive, Germany 17:30 - Lounges on wheels? 14:15 - Levelling up: an HMI roadmap for 15:15 - 3D sensing infrastructure for autonomy. It also assesses the need for, and proposes living aside from our home and our office. Longer Due to upcoming functions on HMI interfaces such Dr Cyriel Diels, academic director, Level 3 and 4 autonomous transport next-generation in-cabin applications possible solutions to, the occupant safety conundrum. distances to commute, more traffic, more time in as augmented reality, 3D displays and autonomous National Transport Design Centre, Derek Viita, senior analyst, in-vehicle UX Dr Gregor O Novak, managing director, the car are already creating this reality today. But driving on SAE Level 3, driver monitoring is Coventry University, UK service, Strategy Analytics, USA NEW Becom Bluetechnix GmbH, Austria 17:00 - Creation and evaluation of how can we avoid this development turning out becoming increasingly important. With C.REF – a Shared and automated mobility may make our What lessons have we learned from rollout of For today’s advanced driver assistance systems and a Level 5 driverless pod design to be our personal mobile dystopia? How do we reference and ground truth block set for test and journeys more pleasurable and productive. Future early semi-autonomous systems? What are best for the next step in autonomous driving, it is important Joscha Wasser, researcher, Coventry create a place to be and a place where we like to validation of several KPIs – CMORE has wide vehicles are envisioned as ‘lounges on wheels’, sporting practices and common pitfalls observed in HMI to know what is going on inside the vehicle. A depth- University & Horiba MIRA, UK be? HMI, AR, Interior Cocoon and new perceptual experience with forward-looking ADAS sensors. flexible seating arrangements with interiors dotted designs of these systems? And what lessons can sensing infrastructure based around time-of-flight Driverless last-mile mobility vehicles, also known as safety features are core to create a Mobile With the new C.REF Gaze and C.REF Head block with screens. This presentation will explore these we apply to improve usability of and encourage trust 3D sensors and VCSEL illuminations combined pods, are on the verge of becoming a reality accessible Livingroom 2.0. These concepts will be introduced, set, two new functions in the reference block sets proposals from a human-centred design perspective in more automated systems? In this presentation with embedded processing provides the necessary to the wider public as part of public transport systems. explained and discussed in this presentation. for interior observation have been developed. and discuss the necessity to respect basic human we look at the current state of on-market semi- sensor data to recognise driver and passengers. However, little is currently known about the passenger These block sets give the development team requirements to realise the potential benefits such autonomous HMI, how this has affected interest Based on the driving mode, increased safety through requirements for these vehicles. We therefore proposed shared and automated vehicles may be able to provide. in and attitudes towards autonomous transport, correct pose detection and a broad array of comfort a comfort model for driverless pods, presented at the and next steps for more automated transport, all functions for all users can be realised. Increased Autonomous Vehicle Interior Design & Technology through the lens of worldwide consumer research. context and user awareness of the system enable Symposium 2017, which was then used as a basis DAY 2 WEDNESDAY 6 JUNE a more intuitive user experience and novel HMIs. for the design of a four-seater driverless pod. A digital 14:45 - Individual differences in trust, prototype and an ergonomic buck were then used Autonomous Vehicle Interior Design & Technology Symposium perception and usage of automated systems 15:45 - 16:30 - Break to conduct participant lead ergonomic evaluations Dr Tanja Schweiger, manager, of the interior and to validate our proposed model. automotive solutions, J.D. Power NEW 16:30 - Occupant restraint dilemma 09:00 - 12:45 - Morning Session Europe GmbH, Germany for autonomous vehicles 17:30 - Virtual prototyping to virtually Recent J.D. Power international studies performed Dr Gopal Krishnan Chinnaswamy, senior test passenger comfort and safety in the USA, China and Germany show many project engineer, Virtual Engineering Caroline Borot, business development 09:00 - Understanding occupants’ visual examples considering potential future market in an ergonomically optimised interior design where cultural differences concerning individuals’ trust Centre, University of Liverpool, UK industry solutions, ESI, France behaviour – AI inside autonomous vehicles dominance by each of the following: automotive OEMs, every other surface is curved. A new type of low- in automated vehicles. A deeper look shows that The Industry 4.0 revolution is enabling the This paper investigates how virtual interior prototyping Modar Alaoui, CEO, Eyeris, USA shared mobility, tech giants and disruptive startups. cost flexible display technology, plastic OLCD, now besides cultural differences there are also many age development of automated driving without interference and digital human models can be used to find the right This session will cover the latest embedded vision Based on the scenarios explored at the Autonomous provides a viable solution for future cars where large and gender differences that should be considered (autonomy). Connected vehicles are expected to car interior design, optimising the comfort of occupants AI technologies that enable visual behaviour Vehicle Interior Design & Technology Symposium area, high brightness and long-lifetime displays when developing automated systems. These be more common in the coming years, and fully and their safety in a completely new, innovative interior understanding for the driver and passengers of in 2017, new approaches and expanded concepts are needed. OLCD uses LCD technology, which is differences concern not only the trust in these autonomous (L5) vehicles are expected within a decade layout. After a brief description of ESI digital thermal autonomous and highly automated vehicles (HAVs). including relevant examples will be presented. The already qualified for automotive displays, with the systems, but also the perception and usage of or two. Autonomous vehicles will require sufficient human models and virtual seat model, it will show Critical to ensuring safety and comfort for all occupants, conclusion drawn is that different markets will see added benefit that it can be conformed and shaped. them, including many interior and HMI aspects. features to ensure the safety of the occupants even through different industrial use cases, how the the intelligent software uses standard cameras different market shares of these four examples, when their positions are not structured as at present. (dis)comfort of the passenger can be virtually tested and to provide emotion recognition from facial micro- depending on the demographic characteristics. This 12:15 - Inducing trust: AI-powered assistants This paper addresses the issues brought about by optimised at the early stage of the car interior design. expressions, 30+ face analytics, body pose tracking, will strongly impact on the user interaction with the as spokespeople for the autonomous vehicle the autonomy revolution and particularly Level 5 action and activity recognition. In the second half of vehicle interior and configurations of each extreme. Dr Nils Lenke, director corporate research, this session, we will cover a number of metrics that Nuance Communications, Germany trigger the activation of support systems, and others 10:30 - 11:15 - Break AI-powered automotive assistants (AA) play an DAY 3 THURSDAY 7 JUNE that enhance the overall ridership experience. increasing role for today’s drivers, enabling them to 11:15 - User-centred design of interact with their vehicles while driving. An interesting 09:30 - A visual aid against motion sickness autonomous driving systems in trucks question is how this will change with the advent of the Dr Paolo Pretto, research team leader, Arnd Engeln, professor for market and advertising autonomous vehicle (AV). One view is that drivers as Max Planck Institute for Biological research, traffic and transport psychology, passengers in AVs will no longer need to use speech. 09:00 - 12:45 - Morning Session Cybernetics, Germany Stuttgart Media University, Germany This paper argues the opposite. Drawing from the Anecdotal evidence indicates that car passengers In the TANGO scientific research project, sponsored literature and new usability studies and research it develop motion sickness when they experience by the German BMWi and in cooperation with Bosch, shows that AAs can play a crucial role in building up 09:00 - Functional interior lighting: light-emitting surfaces for an adaptive UX. And to allow the VI-DAS user to anticipate what’s unexpected and/or sustained discrepancies between MAN, VW and Stuttgart University, we developed an the necessary trust in the users of AVs. It will also show cognition, orientation, wellbeing and safety there are more surprising functional light effects on ahead using contextual sensor data, to support a physical and visual motion cues (e.g. while reading activity and vigilance management system for truck that in AVs there is a practical need to interact with AAs. Dr Frederik Diederichs, senior researcher, humans, as we demonstrated in the luxury Level safe and comfortable user-vehicle collaboration during the travel). This issue is likely to become more drivers using a user-centred design process (UCD). Fraunhofer IAO, Germany NEW 5 Audi Concept Car in the 25th Hour project. for different levels of vehicle automation. deleterious in autonomous vehicles, where passengers We first observed and interviewed truck drivers in 12:45 - 14:15 - Lunch Automotive interior lighting is an emotional feature. may be facing away from the driving direction and their daily work to find out their needs for assistance. Light affects our biological clock and cognition. 09:30 - VI-DAS HMI for user- 10:00 - UX 2025: investigating the user engaged in non-driving activities. We are testing Based on this, we conceptualised innovative ideas to It guides attention, counteracts motion sickness centred automation experience in the connected cockpit of 2025 the efficacy of in-car solutions that will constantly keep drivers active and vigilant in phases of automatic and acts as a natural interface in our mobile smart Paul Schouten, UX designer, TomTom, Patrice Reilhac, innovation and collaboration display to all passengers additional, non-invasive driving on Level 2 and 3. In our further UCD process, home 4.0. Light in the automotive interior can be a Netherlands research director, Valeo, Germany visual information on the current and imminent vehicle these ideas are developed and evaluated in an unique selling point and create corporate identity. The presentation will outline an iterative user- A clear vision of megatrends and technological body motion. The goal is to minimise the perceived iterative process. Interim results will be shown. The fully styled car interior offers best integrated centred design study to find solutions for how progress is fundamental for envisioning the future sensory mismatch and increase passengers’ situational awareness, thus improving safety and comfort. 11:45 - Personalising the autonomous vehicle interior with flexible plastic LCDs 10:00 - Autonomous mobility Simon Jones, commercial BOOK YOUR PASS ONLINE NOW! 10% EARLY-BIRD DISCOUNT 2050: four potential transformative director, FlexEnable, UK 3-DAY PASS €1,950 €1,755 + VAT 2-DAY PASS €1,750 €1,575 + VAT outcomes for vehicle interiors With the gradual transition to autonomous cars, Carlo Budtke, senior consultant, P3 the HMI will have to address the entertainment, Group, Germany information and communication needs of the ONE PASS – ACCESS ALL THREE EVENTS! Carlo Budtke will explore the extremes between passengers. As displays become more and larger, it OVER 150 SPEAKERS PLUS OVER 80 EXHIBITORS different approaches in the industry using four is increasingly difficult to accommodate flat displays

14 www.autonomousvehicleinteriors.com www.autonomousvehicleinteriors.com 15 10% EARLY-BIRD DISCOUNT DAY 3 REGISTER ONLINE NOW! cockpit. Automation Levels 3 and 4, 5G connectivity transitions more intuitive and adaptive. Application 12:15 - VI-DAS Project – a novel approach and AI-enabled adaptive and personalised content examples based on different HMI elements (e.g. to next-generation vehicle interaction allow the future cockpit to be highly focused smart steering wheel) will be presented. Dr Oihana Otaegui, head of ITS and engineering on the human. Valeo presents investigations department, Vicomtech, Spain on its Cockpit 2025 concept for intuitive driving 11:45 - Human and automation as team VI-DAS will progress the design of next-gen 720° and travelling of future highly digital users. members: the AutoMate project connected ADAS (scene analysis, driver status). Dr Andreas Lüdtke, group manager, Advances in sensors, data fusion, machine learning 10:30 - 11:15 - Break Offis, Germany and user feedback provide the capability to better The presentation describes the concept, process understand driver, vehicle and scene context, facilitating 11:15 - HMI for intuitive and and findings of the AutoMate project. The vision of a significant step along the road towards truly semi- 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY adaptive transitions AutoMate is a novel driver-automation interaction autonomous vehicles. Predictions on outcomes in a Dr Alessia Knauss, research specialist, and cooperation concept to ensure that highly scene will be created to determine the best reaction to Autoliv Research, Sweden automated driving systems will reach their full feed to a personalised HMI component that proposes Transitions between automated vehicle and manual potential and can be commercially exploited. optimal behaviour for safety, efficiency and comfort. AI, MACHINE LEARNING AND SOFTWARE driving represent one of the major challenges in This concept is based on viewing and designing autonomous driving. In order for the driver to be the automation as the driver’s transparent and 12:45 - 14:15 - Lunch useful, they should be intuitive and adaptive to the comprehensible cooperative companion or DEVELOPMENT FOR AUTONOMOUS DRIVING driver state, the environmental situation, and personal teammate. This kind of system can enhance safety preferences and characteristics of the driver. This and comfort by using the strength of both the talk will focus on the HMI as an element to make automation and the human driver in a dynamic way.

14:15 - 16:15 - Afternoon Session UPDATED PROGRAMME 〉〉〉

14:15 - Feeling your car: HMIs as new interaction modality is based on cooperation, i.e. driver and the vehicle work together as a team. envisioned by the VI-DAS project mutual support in perception and action between the ADAS&ME is an EU-funded project targeting the Dr Margarita Anastassova, research driver and the car. The cooperation aims to exploit and development of adaptive ADAS, able to decide engineer, CEA LIST, France make concrete the complementarity of the human when and how the vehicle needs to take over or The VI-DAS vision of future in-vehicle HMIs for and the automation as part of a team. The tool able recover control based on driver’s state and the Pass includes Nvidia autonomous and semi-autonomous driving has to allow the cooperation is the HMI. In the project, environmental context. The approach uses adaptive & Siemens workshops a special focus on HMIs providing multi-sensory when the human should compensate a human limit, HMI strategies to assist the driver with automation See page 21 alerts and improving drivers’ situation awareness. An a negotiation-based HMI is implemented to cooperate when needed, and to achieve smooth transitions approach to integrating such HMIs in regular driver with the driver, increasing comfort and acceptability. between automated and manual driving. activity when driving and being driven will be presented. 15:15 - Driver-state-based HMI in automated 14:45 - How can I help my driving: the ADAS&ME approach autonomous vehicle? Stella Nikolaou, researcher, CERTH/ Elisa Landini, programme manager, HIT, Greece RE:Lab, Italy The introduction of automated functions in vehicles This presentation describes the innovative interaction brings a new set of possibilities, but also several paradigm between drivers and highly automated challenges. It is no longer just the driver using the vehicles, developed in the AutoMate EU project. This vehicle as a tool for transportation; instead, the Autonomous Vehicle Software Symposium “Great conferences for international networking and inspiring presentations. The conference meets the trends and is well balanced between future concepts and innovative solutions for current challenges.” Frederik Diederichs, Engineering System / Human Factors Engineering & Vehicle Interaction, Fraunhofer IAO 10% EARLY-BOOKING DISCOUNT! An integral part of Go online to book your pass! Autonomous Vehicle NETWORKING EVENING PARTY TECHNOLOGY Wednesday 6 June, 18:00hrs, Exhibition Hall 5 WORLD EXPO 2018 All delegates and speakers are invited to attend our THEPLUS MOST HMI IMPORTANT CHALLENGES AUTONOMOUS AND NEXT-GENERATION VEHICLE complimentary networking evening. ONBOARDCONFERENCE SERVICES OF ANDTHE YEARTECHNOLOGY

16 www.autonomousvehicleinteriors.com www.autonomousvehicle-software.com 17 10% EARLY-BIRD DISCOUNT DAY 1 REGISTER ONLINE NOW! 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY DAY 2

DAY 1 TUESDAY 5 JUNE 15:45 - 16:30 - Break significantly improve the current methodology conflicting requirements and how to architect your data for training and validation of neural networks. centre to meet future regulations and requirements 16:30 - Efficient training and testing of for capacity, performance, collaboration and growth. autonomous vehicle AI using simulation 17:00 - Preparing for the inevitable 09:00 - 12:45 - Keynote Presentations Rodolphe Tchalekian, EMEA pre- data growth challenges of ADAS 17:30 - Using compilers for sales engineer, ESI GmbH, Germany Larry Vivolo, senior business development safety-critical systems Massive amounts of labelled and unlabelled manager, automotive and electronic Dr Marcel Beemster, CTO, Solid 09:00 - Automated road friction 10:00 - Cognitive IT with storage and 11:45 - Contract-based design of datasets are needed both for training autonomous design automation, Dell EMC, USA Sands BV, Netherlands estimation using car-sensor suite: software defined solutions for ADAS automotive software systems vehicles to navigate complex and unexpected The growing number of high-resolution sensors in Compilers are ‘just’ tools in, for example, the ISO machine learning approach Frank Kraemer, systems architect, Fabio Urciuoli, business development manager, scenarios, and to evaluate that training. As a cars is driving new data management challenges for 26262 functional safety standard for the automotive Mats Jonasson, technical expert, Volvo IBM, Germany Siemens PLM Software, Germany substitute for hours of recordings, Pro-SiVIC ADAS simulation and development. A single sensor industry. Developers prefer to do on-target application Cars, Sweden Advanced driver assistance systems (ADAS/ Organisations have long struggled to break down creates synthetic data to simulate the output from for SAE Level 3 autonomy can consume >4 PB of testing over compiler qualification. However, this does Automotive active safety systems can significantly autonomous driving) are becoming part of all vehicles. large, complex embedded applications into multiple sensor systems for outdoor scenarios storage to simulate 200,000 driven kilometres. not take into account the complexity of a compiler benefit from real-time road friction estimates (RFE) All major OEM and Tier-1 auto manufacturers are manageable isolated components and boost cross- that combine vehicles, obstacles, pedestrians, SAE Level 5 may require 240,000,000km. Legal and the artefacts it introduces into the generated to adapt driving styles specific to the road conditions. implementing and testing AD facilities. We examine functional collaboration. It is clear that automotive weather and road conditions. We will demonstrate obligations may require sensor data archiving for code. Without detailed knowledge of the compilation This work focuses on using the Volvo car-sensor how real-time sensors, big data computing, data storage OEMs, suppliers and involved system designers are how a platform running deep learning algorithms decades, with only days for recovery. Machine process from source code to machine code, it is suite for road friction estimation and prediction and data archiving are integrated in today’s ADAS/ faced with several challenges. We believe that an can receive and process Pro-SiVIC data in learning will set new performance requirements, all incorrect to assume that high code and branch tasks using machine learning algorithms. First, the AD systems, providing a fascinating case study, best architecture-driven development associated with real time. Some application examples will be while IT budgets shrink. During this sessions we will coverage at the application source code translates most significant sensors for RFE estimation are practices for workflow design, testing and development, contract-based design will be helpful for software provided to illustrate how simulation could review how distributed file systems help solve these to similarly high coverage at the machine code. detected by feature ranking, including: ambient data storage and archiving, applicable to all industries. teams to be synchronised from start to finish and to temperature, GPS location, vehicle speed, forces Come hear a fascinating investigation of an industry/ efficiently manage the integration activities from the and road surface, tyre types. Next, image processing technology that will soon affect us all, every day. start. The proposed workflow will provide industrial DAY 2 WEDNESDAY 6 JUNE modules are invoked to detect the drivable surface value impact in terms of early design error detection, condition. Finally, information from processed 10:30 - 11:15 - Break predictable integration of the system, reduced release segments of road images will be fused with vehicle time and flexible organisation level adoption. dynamics sensors to initiate RFE-related warnings. 11:15 - Accelerating connected and autonomous vehicles through 12:15 - AUTOSAR adaptive platform 09:00 - 12:45 - Morning Session 09:30 - Artificial intelligence open-source software for intelligent vehicles in the driver’s seat Dan Cauchy, executive director, Automotive Grade Dr Thomas Scharnhorst, spokesperson, AUTOSAR Serkan Arslan, director of automotive, Linux, The Linux Foundation, USA Development Partnership, Germany 09:00 - Deep learning on Hadoop thinking (after learning and spending significant art machine learning approaches, e.g. measuring Nvidia EMEA, Germany The race to roll out new technology features and AUTOSAR Classic was released more than 10 Dr Tobias Abthoff, member of the energy). Nobel Prize winner Daniel Kahneman calls F1 score, are inadequate to measure failure rates Artificial intelligence is transforming every industry, autonomous vehicles continues to heat up. To compete years ago, with the first release entirely intended executive board, NorCom Information these two different processes fast thinking (System of a few in a billion; a 99.999% F1 score is not but perhaps none more than transportation. From at the speed of a tech company, many auto makers for the embedded architectures of classical ECUs. Technology AG, Germany 1) and slow thinking (System 2). Right now, most considered meaningful. We describe an approach for how we interact with vehicles, to how they drive us, have shifted from traditional development processes AUTOSAR has now developed a completely new The software stack of autonomous vehicles of the attention in autonomous car development is safety-reinforced training, and analysis of perception to how our city’s infrastructure will automatically to agile, rapid development through open-source approach – AUTOSAR Adaptive Platform – to cope increasingly consists of deep learning networks in using System 2 thinking. AI/machine learning is and decision components using simulation. adjust to reduce traffic congestion, deep learning will software. Dan Cauchy will provide an overview of with the challenging environment of internet access complementing traditional software. The development an excellent parallel to the neocortex of the human play a vital role. Nvidia is building AI platforms for AGL, key milestones and the project roadmap. He in cars to make vehicles intelligent and adaptive. This and particularly the verification of such networks brain. We have developed the artificial amygdala 12:15 - Operating and optimising autonomous autonomous vehicles as well as smart cities. The talk will also discuss AGL’s vision for functional safety as aims to support dynamic deployment of customer requires completely new paradigms. We will present a and will give a live demonstration of how it works. vehicle fleets – distributed cloud platform will provide insight into and showcase demonstrations well as for an open-source platform for autonomous applications in providing an environment that new way of efficiently training and verifying deep neural Zhao Lu, CTO, BestMile, Switzerland of the existing state of autonomous vehicles in driving that will help accelerate the development of requires high-end computing power and connects networks by combining deep learning with state-of-the- 10:30 - 11:15 - Break The benefits of autonomous vehicles can only be development and in production, as well as discuss self-driving technology while creating a sustainable deeply embedded and non-AUTOSAR systems art big data technology. The approach targets globally leveraged when integrated into a coherent and the steps to fully autonomous Level 5 robo-taxis. ecosystem that can maintain it as it evolves over time. in a smooth way while preserving typical features distributed teams in particular, and also distributed 11:15 - Autonomous fleet management coordinated mobility system. Challenges lie in originated in deeply embedded systems like safety. non-movable test data. The general architecture will — challenges and opportunities how mobility providers will be able to offer services be presented and then an actual use case in the field Pejvan Beigui, CTO, EasyMile, France with autonomous vehicles when they can no 12:45 - 14:15 - Lunch of image understanding will be shown in detail. Fleet management is a key component of the coming longer rely on drivers and current software is not autonomous vehicle-based mobility as a service sufficient. There is a clear need for a platform to 09:30 - Making cameras self- revolution. At EasyMile, we are the centre of this create coordinated, efficient, flexible and sustainable 14:15 - 18:00 - Afternoon Session aware for autonomous driving revolution as we have been designing both the EZ-10 mobility services in which autonomous vehicles Autonomous Vehicle Software Symposium Dr Florian Baumann, technical director, autonomous shuttle and the software stack required are operated and optimised as a fleet, meeting Adasens Automotive GmbH, Germany for autonomous driving. We have helped transport real-time demand or adhering to a schedule while 14:15 - Virtual validation techniques 14:45 - Challenges of neural 15:15 - Complex deep learning software Two fundamental algorithms addressing the issue operators around the world with operating fleets of adapting to network disruptions. The paper will for CNN-based ADAS/AD systems networks for vehicle software stacks – revealing their inner secrets of making cameras self-aware of their status are EZ-10. Our fleet management solution has been detail the technical specificities of such a platform. Vignesh Radhakrishnan, senior ADAS/AD Joshua Davis, software engineer, Illya Rudkin, principal software engineer/ proposed: online targetless calibration based on architected for high availability, fault tolerance and systems engineer, AVL, UK Horiba MIRA, UK safety-critical software development optical flow, and blockage detection based on image high scalability, while maintaining other key properties 12:45 - 14:15 - Lunch Virtual validation will become an important aspect of In recent years, artificial neural networks have been lead, Codeplay Ltd, UK quality metrics (e.g. sharpness and saturation). such as cybersecurity and agility. In this talk, we ADAS/AD development. However, there is an urgent shown to be very effective at learning high-level The integration of complex software with hardware The online calibration is based on the vanishing will discuss the challenges we’ve faced building need to define techniques for validating AI-based semantic information from unstructured data such as while meeting tight development constraints is a point theory; the soil/blockage detection is based this system, and present some of our key results. ADAS/AD systems. Due to the stochastic nature of AI audio, image and video, modelling complex non-linear challenge for all automotive companies. For functional on the extraction of image quality metrics and systems, the challenge is to make sure the systems relationships between input and output signals and safety engineers the scope of concerns with an the identification of discriminative feature vectors 11:45 - Safety-reinforced AI act as per defined requirements. AVL is leading a allowing us to make sense of the world around us. This expanding code base supporting diverse hardware by a support-vector machine. The presentation driver development research project – SAVVY (Smart ADAS Verification paper will give a brief introduction to neural networks, is immense. Khronos moves open standards APIs will include videos and real-world examples Dr Edward Schwalb, lead scientist, MSC & Validation Methodology) – funded by Innovate discuss some of the big challenges that we have seen such as OpenCL and OpenVX to be ISO 26262 from the algorithms running in real time. Software, USA UK, which is addressing the challenge of defining a putting them to use in autonomous vehicle applications compatible software/hardware enablers. Codeplay Safety must be a major focus for automated vehicle process to validate CNN-based ADAS/AD systems. such as training and validation, and review some of the will show, using Tensorflow as a use case, how a 10:00 - The amygdala of the self-driving car (AV) development. Today humans drive 100 million methods being used to overcome these challenges. holistic toolchain with open standards can manage Raul Bravo, CEO, Dibotics, France miles between fatal crashes. Consequently, proving complex software stacks using SYCL and OpenCL. When using the human brain as a model for that a specific revision of an AI driver is safe is This helps to enable developers and safety functional intelligence, the human amygdala allows for not practical using road and track tests; neither engineers to manage issues such as power usage and ultra-fast, deterministic and effortless (low-power) investigations of Tesla’s crash nor Waymo driving concurrency timing while mitigating safety concerns. reaction, while the neocortex allows for complex records resulted in conclusive results. State-of-the-

18 www.autonomousvehicle-software.com www.autonomousvehicle-software.com 19 10% EARLY-BIRD DISCOUNT DAY 2 REGISTER ONLINE NOW! EXCLUSIVEDAY DELEGATE WORKSHOPS 7 JUNE 2018 DAY 3

14:15 - 18:00 - Afternoon Session NEW FOR 2018! 14:15 - Producing systems that applied to the development of autonomous driving 17:00 - Updating autonomous enable the innovation that vehicle functions. The different modelling layers will vehicle software remotely – even EXCLUSIVE HANDS-ON WORKSHOPS FOR ACCELERATING AI AND DEEP autonomous vehicles will require be explained, and the benefits of the approach will be over non-secure channels Agustin Benito Bethencourt, principal discussed and illustrated through practical examples. Alberto Troia, memory system architect, Micron LEARNING AND END-TO-END DEVELOPMENT OF AUTONOMOUS VEHICLES consultant - FOSS, Codethink Ltd, Spain Technology Inc, Germany In order for autonomous vehicles to react to new 15:15 - Building artificial brains that learn For autonomous driving to become mainstream, 10:30 - 12:00 - Morning Workshop 14:15 - 16:15 - Afternoon Workshop situations and demands, software and data will need to driving better and more quickly than humans passengers must trust the autonomous vehicle be updated on a regular basis, which requires the way Karim Mansour, vice president and co-founder, enough to give up driving control. As a broader in which software systems are produced today to be Sigra Technologies GmbH, Germany component of that trust, it will also be essential for turned upside down. With Open Source best practices This talk will concentrate on how artificial intelligence OEMs to implement an infrastructure to securely and agile principles in mind, along with a background can help build sophisticated brains that can drive in support software updates over remote, non-secure in other industries, Agustin will go over some of the complex scenarios. It will look at how these brains can channels as Secure Over-The-Air (SOTA) software key changes that auto makers and Tier 1s will need react much more quickly than human brains, and how updates become mainstream. In this paper, we will to face in the near future to enable all that innovation they can detect future hazards using simple data that present a methodology based on remote diagnostic in vehicles in a sustainable way. He will focus on human beings can simply ignore or can’t detect or see. software technologies to establish the authenticity The Nvidia Deep Learning Institute (DLI) offers We are delighted to announce that Siemens will be delivery and maintenance processes and practices. of the vehicle and associated software updates. This hands-on training for developers, data scientists and hosting a world-first workshop in Stuttgart this year. 15:45 - 16:30 - Break methodology will prohibit cyberattacks when software 14:45 - Autonomous vehicles: NEW is being downloaded or uploaded from the vehicle. researchers looking to solve challenging problems Co-hosted with Mentor and TASS International, providing software features quickly 16:30 - Running a functional and by model-based system design open HAD software architecture on an 17:30 - The zero-defect software with deep learning and accelerated computing. this session will provide insight on the end-to-end Sébastien Christiaens, department manager, FEV Adaptive AUTOSAR infrastructure factory – myth or reality? development of autonomous vehicles, with subjects Europe GmbH, Germany Rudolf Grave, head of product systems architecture, Ingo Nickles, senior field application engineer, Join us for a hands-on DLI workshop called Introduction to Autonomous driving vehicles are complex systems. Elektrobit Automotive GmbH, Germany Vector Software, Germany Object Detection with TensorFlow. This workshop is a lightning including: Existing processes for component-orientated This talk will explain how functional software Software development can no longer be considered introduction to object detection and image segmentation for data development reach a limit. New approaches are architecture with open interfaces and software modules as a creative process leading to mastery of how the • Model-based development, verification and validation required to provide safe and affordable solutions. can be integrated on a high-performance micro- function and logic is expressed in the application’s scientists, engineers and technical professionals. This task of framework for automated vehicles System modelling approaches offer the opportunity controller using Adaptive AUTOSAR middleware. In lines of code. Because the role of software is to computer-based image understanding permeates many major to smooth the steps from customer requirements addition to the functional challenges, the handling of deliver key features and safety-critical functions in to software development, and enable reuse and automotive safety integrity levels will be shown. The so many pieces of equipment and systems, it now fields such as advertising, smart cities, healthcare, national • Framework consisting of advanced simulation environments front-loading, leading to considerable effort and benefit of an open software framework for automated needs to be constructed with the precision and quality defence, robotics and autonomous driving. Ultimately, the time reduction for integration and testing. This driving combined with a dependable operating that can be seen in any modern manufacturing (MiL, SiL, HiL, ViL) as well as physical testing facilities presentation shows how a systematic approach to environment is increased time to market due to fast process. This presentation will discuss whether a goals of this course are to provide a broad context and clear (laboratories, test tracks, public roads) system requirements definition can practically be integration and early testing on a system level. zero-defect software factory can be a reality. roadmap from traditional computer vision techniques to the most recent state-of-the-art methods based on deep learning • Facilitating the full spectrum of automated driving technology, DAY 3 THURSDAY 7 JUNE and convolution neural networks (CNNs). ranging from system-on-a-chip design, sensor development and systems integration to full-vehicle performance evaluation If you do not have any experience with deep learning, we and traffic impact analysis recommend you take at least the Image Classification with 09:00 - 10:00 - Morning Session DIGITS lab from www.nvidia.co.uk/dlilabs prior to attending. All delegates are welcome to attend this exclusive workshop. The Siemens team are focused on delivering an educational 09:00 - Hive mind: AI and NEW tasks in the cloud. Attendees will learn how to leverage the world have collected terabytes of raw sensor data platform for participants that will result in a more streamlined, Autonomous Vehicle Software Symposium ML for vehicle and fleet both edge compute and a shared AI/ML platform and must now sift through that data to uncover the robust and faster automated vehicle development process. Herman Coomans, senior solutions architecture without having to build the underlying IT infrastructure. moments that capture these rare objects and scenarios. Access to both manager, Amazon Web Services, Australia In this presentation, we discuss the challenges behind Machine learning models require massive compute 09:30 - Only as good as your data developing a system to sort through this immense workshops is resources for training, but can be deployed with much Sheikh Shuvo, product and solutions NEW amount of data, and how to label that data accurately included in the more modest compute in-vehicle. Edge compute with manager, Mighty Ai, USA and at scale. Additionally, we examine best practices machine learning can be used for fast decision making To achieve Level 5 autonomy, vehicles must be able for using manual annotation when accuracy is critical. delegate in the field, and vehicle fleet data, maps and other to process and respond to so-called ‘edge cases’ that pass fee! sources can be used for navigation and fleet training rarely occur and are hard to account for. Teams around

NETWORKING EVENING PARTY Wednesday 6 June, 18:00hrs, Exhibition Hall 5 All delegates and speakers are invited to attend our complimentary networking evening.

20 www.autonomousvehicle-software.com www.autonomousvehicle-software.com 21 LOOK WHO IS SPEAKING... KEY COMPANIES INCLUDE: 5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY

EVENT INFORMATION

STUTTGART

Messe Stuttgart Due to its location between the airport and motorway, Messe Stuttgart Trade Fair Centre has an incomparable traffic infrastructure. Direct connections to the A8 motorway, the B27 trunk road, the airport, the S-Bahn (rapid-transit railway) – and, in future, the planned railway station for local and long-distance trains – ensure that visitors and exhibitors will be able to travel quickly and easily to the Trade Fair Centre. The journey times to a large number of hotels are also surprisingly short. Stuttgart city centre can be reached in around 20 minutes by S-Bahn.

The City Stuttgart is a city with many varied attractions, Porsche and Mercedes-Benz museums, a world-renowned ballet, excellent cultural and sporting highlights, diverse leisure and accommodation possibilities and an international variety theatre.

CONFERENCE PACKAGE INCLUDES

• Access to all conferences at the Autonomous Vehicle Test • Assorted complimentary refreshments during the & Development Symposium 2018 conference networking breaks • Invitation to the drinks and networking reception evening • Access to a secure website containing all presentations • Event app to arrange your individual conference schedule made during the conference (subject to speaker approval) and meetings • Visitor Meetings & Relaxation Area for sit-down discussions • Pre-conference coffee on arrival • Free-of-charge cloakroom • Lunch and refreshments at Conference Dining • Complimentary wi-fi

CONFERENCE RATES HOTELS & TRAVEL 10% EARLY-BOOKING DISCOUNT Book directly through our official partner, RAI Hotel Services, 3-day pass €1,950 €1,755 + VAT so you can easily find accommodation that meets your requirements and suits your budget. 2-day pass €1,750 €1,575 + VAT • Profit from the best rates • No administration or handling fees Group booking discount • Wide selection of hotels Receive an extra 10% discount on each registration for a group You can reserve your hotels directly using the RAI’s online reservation booking (2+ delegates) by making them on the same date, from system (via booking.com). Just enter your arrival and departure dates and the same company. make your choice from a wide selection of hotels. We advise you to book as soon as possible so we can guarantee the finest accommodation at OPENING TIMES affordable rates. See the website for details

Tuesday 5 June Networking Breakfast 08:15 – 08:50 CONTACT US Conference Opening Hours 09:00 – 18:00 Wednesday 6 June Andrew Boakes, UKi Media & Events, Abinger House, Conference Opening Hours 09:00 – 18:00 Church Street, Dorking, Surrey, RH4 1DF, UK Thursday 7 June Email: [email protected] Conference Opening Hours 09:00 – 16:15 Tel: +44 1306 743744 GO ONLINE TO VIEW THE FULL LIST OF SPEAKERS www.autonomousvehiclesymposium.com www.autonomousvehiclesymposium.com 23 THE EXHIBITION – 80+ EXHIBITORS EXPECTED! LOOK WHO IS EXHIBITING... COMPANIES INCLUDE:

Autonomous Vehicle TECHNOLOGY WORLD EXPO 2018

5, 6, 7 JUNE 2018, MESSE STUTTGART, GERMANY

If you are working on an autonomous vehicle project, you need to attend this event!

Your pass includes entry to the exhibition featuring autonomous vehicle design, technologies, development and validation

The Autonomous Vehicle Technology World Expo encompasses three conferences that are dedicated to autonomous vehicle design, engineering, possibilities and validation. The event features a number of unique exhibits from pioneering companies who are focused on autonomous vehicle technologies, and is held alongside Automotive Testing Expo Europe, Automotive Interiors Expo, Engine Expo, and Global Automotive Components and Suppliers Expo, with 800+ exhibits. In short, if you are working on an autonomous vehicle project, you need to attend this event. © Continental

VISIT THE WEBSITE

Workshop Partner FOR THE LATEST UPDATES

www.autonomousvehicletechnologyworldexpo.com www.autonomousvehicletechnologyworldexpo.com MORE REASONS TO ATTEND THAN EVER BEFORE OUR BIGGEST EVENT TO DATE! 10% EARLY-BOOKING DISCOUNT! REGISTER ONLINE NOW!

Event sponsor FOR MORE INFORMATION Contact: Andrew Boakes | UKi Media & Events, Abinger House, Church Street, Dorking, Surrey, RH4 1DF, UK Tel: +44 1306 743744 | Email: [email protected]