Infrastructure Owner Operators Guiding Principles for Connected Infrastructure Supporting Cooperative Automated Transportation

Supporting Technical Concepts February 2020 AASHTO/ITE/ITSA Joint Task Force for IOO Guiding Principles for Connected Infrastructure for Cooperative Automated Transportation

Executive Oversight Jennifer Cohan, Delaware DOT Roger Millar, Washington State DOT

Joint Task Force Co-Chairs Collin Castle, Michigan DOT Faisal Saleem, Maricopa County DOT

Joint Task Force Members AASHTO Tracy Larkin Thomason, Nevada DOT John Hibbard, Georgia DOT Blaine Leonard, Utah DOT Joseph Sagal, Maryland DOT

ITE Steve Kuciemba, WSP Raj Ponnaluri, Florida DOT

ITSA John Kenney, Toyota

Association Staff

AASHTO Gummada Murthy Venkat Nallamothu Matt Hardy Patrick Zelinski

ITE Jeff Lindley Siva Narla

ITSA Amy Ford Steven Bayless

Consultant Support Kyle Garrett, Synesis Partners

ii Terminology

AAMVA American Association of Motor Vehicle LIDAR Light Detection and Ranging Administrators MaaS Mobility as a Service AASHTO American Association of State Highway MOD Mobility on Demand and Transportation Officials MUTCD Manual on Uniform Traffic Control Devices ADS Automated Driving System NACo National Association of Counties AI Artificial Intelligence NCHRP National Cooperative Highway Research Program ATCMTD Advanced Transportation and Congestion Management Technologies Deployment NCSL National Conference of State Legislatures

AV Automated Vehicle NHTSA National Highway Traffic Safety Administration

CAMP Collision Avoidance Metrics Partnership NSTC National Science and Technology Council

CAT Cooperative Automated Transportation NTCIP National Transportation Communications for ITS Protocol CAV Connected and Automated Vehicles OBU On-Board Unit CTSO (AASHTO) Committee on Transportation System Operations ODD Operational Design Domain

CV Connected Vehicle OEM Original Equipment Manufacturer

DAVI Data for Automated Vehicle Integration RSU Roadside Unit

DOT Department of Transportation RLVW Red Light Violation Warning

DSRC Dedicated Short-Range Communication RSZW Reduced Speed Zone Warning

FCC Federal Communications Commission SAE Previously, Society of Automotive Engineers

FHWA Federal Highway Administration SDO Standards Development Organization

GHz Gigahertz SPaT Signal Phase and Timing

GNSS Global Navigation Satellite System TMDD Traffic Management Data Dictionary

GP Guiding Principle TNC Transportation Network Company

GPS Global Positioning System TSMO Transportation System Management and Operations

HMI Human Machine Interface UAV Unmanned Aerial Vehicle

IHS Interstate Highway System USDOT United States Department of Transportation

IOO Infrastructure Owner/Operator V2I Vehicle-to-Infrastructure

ISO International Organization for Standardization V2V Vehicle-to-Vehicle

ITE Institute of Transportation Engineers V2X Vehicle-to-Everything

ITS Intelligent Transportation Systems VSL Variable Speed Limit

ITSA Intelligent Transportation Society of America WG Working Group

iii iv Table of Contents

Executive Summary 1 Introduction and Purpose 2 Overview of CAT 2 Stakeholders and Their Objectives 3 Applicable Modes 4 Vehicle Automation 4 Roadway Automation 5 Technology and Communications 5 Applications 7 IOO Guiding Principles for CAT Infrastructure 8 The Need and Basis for GPs 8 Objectives of the GPs 9 GPs and Concepts 9 Automation 9 Data 12 Telecommunications 15 Operations 17

Collaboration 19 Applying the CAT Infrastructure GPs 20 CAT and IOO Processes 20 Preparing for CAT Infrastructure 22 Future Efforts 23 Resources and References 24 CAT Coalition Activities and Resources 24 National Cooperative Highway Research Program (NCHRP) Studies Related to CAT 24 USDOT/FHWA AV Resources and Projects 24 State Regulatory Perspectives 25 Applicable Standards 25

v Guiding Principles for Connected Infrastructure Supporting Cooperative Automated Transportation

GP1—Automation: Support increased vehicle automation to improve traveler safety, mobility, equity, and efficiency.

GP2—Data: Achieve a connected vehicle ecosystem that enables reliable, secure V2I data exchanges in order to support cooperative automated transportation to improve traveler safety, mobility, equity, and efficiency.

GP3—Telecommunications: Protect and utilize the 5.9 Gigahertz (GHz) spectrum designated for “operations related to the improvement of traffic flow, traffic safety, and other intelligent transportation service applications.”

GP4—Operations: Develop CAT strategies that enhance existing transportation system operational capabilities to improve traveler safety, mobility, equity, and efficiency.

GP5—Collaboration: Collaborate and communicate with OEMs and mobility service providers in the planning, testing, and demonstrations of CAT applications to support eventual interoperability and to achieve positive impacts on safety, mobility, equity, and efficiency.

vi Executive Summary

Cooperative Automated Transportation (CAT) envisions all stakeholders and elements of the transportation system working together to improve safety, mobility, equity, and operations efficiency through interdependent vehicle, infrastructure, and systems automation enabled by connectivity and information exchange. The concept is intentionally expansive, building on work being done among public agencies, industry, and academia to develop connected and automated vehicles and infrastructure systems to support them. It looks beyond existing, developing, and planned transportation concepts to a fully integrated system serving travelers, goods, and services. Automated vehicles and the CAT concept are developing so quickly that the American Association of State Highway and Transportation Officials (AASHTO), Institute of Transportation Engineers (ITE), and the Intel- ligent Transportation Society of America (ITS America)formed a joint Task Force in 2019 to develop guiding principles for infrastructure owner/operators (IOOs) in supporting CAT. The resulting Guiding Principles (GPs) were then reviewed and approved within the three associations, with the final guiding principles being adopted by AASHTO at its Annual Meeting in October 2019. The GPs address five dimensions of IOO development of CAT. The functional core of CAT isAutomation of vehicles and the infrastructure that vehicles share with other CAT users. Data enables the automation. The vehicles and infrastructure may be the devices being automated, but data powers the automation. Communications enable the data interactions between the vehicles, infrastructure, and users. Operations capture and enact decisions about how the transportation system is automated. Collaboration creates an environment that values and incorporates the needs and objectives of all CAT participants. These five dimensions will work together to provide a seamless, consistent, and integrated CAT ecosystem across the nation. GPs are likely to change over time as new information and innovations come along. In the near term, the GPs reflect the consensus direction of the IOOs and can support impact assessment of CAV developments in a rapidly changing CAT ecosystem. The GPs are not intended to constrain CAT deployment. Over the longer term, GPs are intended to give IOOs maximum institutional flexibility while working together to develop and deploy CAT strategy, standards, infrastructure, telecommunications, data exchange, best practices, and public information. This Supporting Technical Concepts document recognizes and acknowledges that agencies, their industry partners, and associations might benefit from additional context to supplement the guiding principles themselves. It is not intended to be a “final” or “complete” description of technical resources that will be needed to support CAT deployments. It does provide a starting point for what will be a continuous process of assessing and responding to evolving institutional goals, operational objectives, and technological developments for CAT. The joint task force will be monitoring developments and amending these CAT technical concepts as appropriate, likely evolving into more formal technical support and exchange efforts.

1 Introduction and Purpose

Cooperative Automated Transportation (CAT) envisions all stakeholders and elements of the transportation system working together to improve safety, mobility, equity, and operations efficiency through interdependent vehicle, infrastructure, and systems automation enabled by connectivity and information exchange. The concept is intentionally expansive. It looks beyond existing, developing, and planned transportation concepts to a fully integrated system serving travelers, goods, and services. CAT includes all modes, systems, and uses of surface transportation, whether by light or commercial vehicle, mass transit, shared mobility service, or bicycle or scooter. CAT is automated in vehicles, infrastructure, and operations—in traffic management, intersection safety, fare collection, mobility services, trip planning and more. CAT is cooperative across the public and private sectors, policies, modes, and services. It requires reliable, low-latency, high-bandwidth communications and two-way data and information exchange among all users, managers, and operators. While the media have focused on development of automated vehicles, those who work in the transportation industries understand that vehicles need infrastructure, and particularly connected infrastructure, to maximize their potential for enhancing safety, mobility, equity, and operational efficiency. The Infrastructure Owner/Op- erators (IOOs) have the complementary opportunity and bear the responsibility to enable the benefits of CAT through connected infrastructure deployment. The purpose of Guiding Principles for CAT Connected Infrastructure is to establish criteria for IOOs to advance connected infrastructure, data, management, and operations supporting CAT solutions. Through collaboration among the American Association of State Highway and Transportation Officials (AASHTO), the Institute of Transportation Engineers (ITE), and the Intelligent Transportation Society of America (ITSA), these principles will encourage interoperability and consistency in CAT deployments and enable IOOs to consolidate and communicate their intentions and the value this represents in effectively managing mobility and safely operating our transportation system. The GPs will enable IOOs to facilitate collaboration, educate the workforce, support interoperable deployments, and inform the public. The GPs themselves are likely to change over time as new information and innovations come along. In the near term, the GPs reflect the consensus direction of the IOOs and can support impact assessment of CAV developments in a rapidly changing CAT ecosystem. The GPs are not intended to constrain CAT deployment. Over the longer term, GPs are intended to give IOOs maximum institutional flexibility while working together to develop and deploy CAT strategy, standards, infrastructure, telecommunications, data exchange, best practices, and public information.

Overview of CAT

A case for and description of connected infrastructure for CAT first needs to lay out a vision for the CAT ecosystem. In many ways, CAT represents simply the next stage in the long arc of surface transportation innovations,- from local and intercity roads, to a national highway network, to the Interstate Highway System (IHS),from no controls, to the first traffic signals, to Intelligent Transportation Systems (ITS).

2 CAT is also however an inflection point in surface transportation. Traffic fatalities across the country are un- conscionably high despite decades of new safety technologies and public awareness. Demand for transportation increases faster than infrastructure deployment alone can satisfy. Mobility suffers as congestion exceeds design capacity and operational controls. Societal costs of crashes, lost mobility, and system maintenance increase faster than system capacity and resources. In the midst of these challenges, CAT recognizes that vehicles and users can evolve from demand producers to become part of the solution to surface transportation challenges. With these concepts in mind, following is a sketch of the CAT ecosystem—its stakeholders, modes, levels of automation, technologies, and applications.

Stakeholders and Their Objectives

Virtually everyone who lives, works, and moves through the world is a CAT stakeholder. But that needs to be broken down to get a necessary set of perspectives on what a CAT ecosystem would look like. Categories are not necessarily exclusive; an agency may act in multiple roles. Infrastructure Owners and Operators (IOOs) manage the roadways on which people, goods, and services move. They typically also have planned, designed, and built those roadways. IOOs are generally, but not al- ways, public agencies working for a state, local, or tribal government. They are responsible for providing safe and efficient transportation services for their citizens and constituents. Motor Vehicle Regulators and Administrators license the drivers and vehicles operating on the roadways on behalf of state, local, and tribal governments. They assure that the system drivers and vehicles meet a minimum set of requirements for safe operations and behaviors. Law Enforcement Agencies monitor vehicles and drivers on the roadways to assure that they comply with traffic controls put in place by the government and IOOs. Law Enforcement also works withEmergency Services and IOOs to respond to and restore safe operations following crashes and other incidents. Vehicle Operators are primarily the drivers of the vehicles, but may indirectly include fleet operators and dispatch services for Freight Operations, Transit, and Mobility Service Providers.1 These stakeholders are primarily concerned with safely and efficiently getting from one point to another across the road network. Their aggregate behaviors are also the primary contributors with the IOOs in determining the relative safety and efficiency of the network as a whole. Pedestrians and Non-Motorized Vehicle Operators are the most vulnerable users of the roadways. Their safety is the most at-risk in any conflict with vehicles, and they typically have the least control over their circumstances interacting with the infrastructure. Their safety and mobility are key indicators of the relative equity of CAT. Vehicle Manufacturers (or, Original Equipment Manufacturers, OEMs) and Automated Driving System (ADS) Developers provide vehicles and automation software for moving people, goods, and services across the roadway network. They have been working with IOOs and communications and technology providers to develop CAT applications and implement standards.

1 Mobility Service Providers are alternatively referred to as Transportation Network Companies (TNCs). The former term is used throughout this document to recognize that not all mobility service providers might be companies and to reduce potential confusion with the transportation (or roadway) network itself.

3 Communications and Technology Providers supply IOOs and vehicle manufacturers with equipment and services supporting CAT applications. In this role, they are primary implementers of communications and data standards. Transportation Information Service Providers collect data from both vehicles and infrastructure to synthe- size new data sets supporting safe and efficient transportation. Their clients may be any combination of other stakeholders.

Applicable Modes

A fully deployed CAT ecosystem will include all the types of vehicles and modes that operate on the infrastructure, with varying degrees of cooperation, automation, and integration. Light passenger vehicles, including light trucks, will likely continue to be the majority of vehicles in most U.S. settings. Vehicle configurations may become more diverse over time, but vehicles on the road today will still be part of the light vehicle fleet in 20, or even 30 years. Freight vehicles have seen some of the earliest automation technology demonstrations and are likely to adopt full automation in some operational design domains (ODDs) before light passenger vehicles. Transit vehicles benefit from having relatively predictable routes and limited ODDs, to the extent of seeing full automation in some settings like campus and downtown shuttles. Emergency services vehicles and roadway maintenance vehicles face ODDs and operational requirements beyond other light and large vehicles, and may be later to achieve full automation. They are nonetheless a crit- ical part of the CAT roadway ecosystem that will be early significant contributors of roadway and event data to be shared with other vehicles, and will enable much safer and efficient emergency response. Micromobility vehicles—scooters, ebikes, and such—are an important emerging category that is seeing significant innovation and growth of use. It is not clear how much automation will be available to micromobility vehicles, but the category could include automated package delivery devices, low-altitude unmanned aerial vehicles (UAVs), and other non-traditional vehicles.

Vehicle Automation

The concept of vehicle automation intends to eventually enable an automated driving system (ADS) to perform “all aspects of the dynamic driving task under all roadway and environmental conditions that can be managed by a human driver”.2 The concept is extensible across vehicle classifications, powertrains, and modes, even to micromobiles. SAE International has developed descriptions of levels of automation from 0 (no automation) to 5 (full automation). Full vehicle automation will develop over time in functional increments as the ADS take on increas- ingly complex tasks and ODDs. A fully cooperative automated transportation system will leverage widespread deployment of automated vehicles and enabling roadway automation. In the meantime, vehicle automation will develop in a mixed environment of varying levels of automation and cooperative capability.

2 SAE International. J3016_201806: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles (Warrendale: SAE International, 15 June 2018). https://www.sae.org/standards/content/j3016_201806/, accessed January 27, 2020.

4 Roadway Automation

Although the focus of popular media is on automated vehicles, roadway automation has been developing since at least the first traffic signal systems were deployed in the early 20th century to address intersection safety and congestion. Intelligent Transportation Systems (ITS) in the late 20th century developed alongside the enabling computing and communications technologies to inform and support operational decision making. ITS greatly expanded capabilities for transportation systems management and operations (TSMO), providing IOOs with an ability to influence traffic conditions beyond the deployment of new hard infrastructure. The advent of connected and automated vehicles provides new channels for roadway automation concepts to develop. CAVs will empower new capabilities in their capacity to provide very granular data—down to the level of individual vehicle experience—for seeing traffic and roadway conditions. The aggregation and processing of these data will provide a much broader and more detailed view of conditions across the transportation network than SAE International Levels of Automation has previously been available to IOOs. This more accu- rate view of conditions enables better decision making at a system level, with more precise traveler information and traffic controls deployed on the roadways. Human drivers and ADS will then be able to make better decisions at a finer level of control than has previously been possible.

Technology and Communications

Building a CAT ecosystem involves the same concepts and components as are needed in ITS and CV applications, with the addition of automation components on vehicles and across the infrastructure. The components are distributed among and across vehicles, the roadside, infrastructure services, and external information services as described below. Communications are the essential means of cooperation in CAT. All of the components within any of the four component domains operate within their own networks. Interactions between the domains may, however, use media and protocols particular to those interactions. Vehicular components supporting CAT include the automated driving system, operating alongside or in lieu of the human driver, the vehicle sensors, and the human–machine interface. Vehicle sensors could include positioning and detection systems using a global navigation satellite system (GNSS) such as the U.S. Global Positioning System (GPS), optical cameras, and light detection and ranging (LIDAR). An on-board unit (OBU) provides the interface for communications with other vehicles and the network. Traveler information and entertainment applications (whether built-in or on personal mobile devices) in the vehicle will use cellular network services or Wi-Fi (where available).

5 Interactions between vehicles and roadside devices will use specific vehicle-to-everything (V2X) wireless communications and protocols. Dedicated Short-Range Communications (DSRC) was developed specifically for this purpose and has been used in CV application development and deployment since 2003. An alternative technology, cellular V2X, or C-V2X, has been developed to be deployed as part of a cellular environment. The roadside is the IOOs’ domain, and devices needed there for CAT are extensions of those used in ITS and CV applications. Broadly categorized, these would include sensors for traffic and weather, controls for signals and gates, and displays for traveler information, speed limits, and other local traffic control signs. A CAT ecosystem might deploy new application-specific automation services—perhaps variable speed limits based on congestion—to a roadside field controller. A roadside unit (RSU) using V2X communications provides connections to vehicles, and a backhaul interface connects the roadside to other IOO systems, such as a traffic management center. Components within the IOO network for CAT will build off existing traffic, maintenance, and other management systems to provide real-time network status awareness and automated response to incidents, weather, and other events. Responses will be based on and similar to TSMO strategies deployed in non-automated contexts, and will likely use machine learning, artificial intelligence (AI), and real-time and predictive analytics to identify network states and mitigations. These algorithms may also use third-party, cloud-based services to supplement the IOO’s own data and methods. Vehicles and IOO systems in a CAT ecosystem will also depend on third-party data and services. Much of the security framework, for example, will depend on certification and credentialing services that are, by necessity, provided by trusted third-party authorities. Communications, device, vehicle, and driving system vendors will use those services in providing updates to their devices. Cloud-based services with which IOOs may interact are likely to include map providers, weather data, and public and commercial data aggregators. Vehicles may be interacting with other traffic data, entertainment, and advertising services.

6 The CAT Ecosystem

Applications

Application use cases provide a view into how the needs, concepts, technology and operations would come together in a CAT ecosystem. Most CAT applications will build on ITS and TSMO experience, and on the early connected vehicle (CV) applications. From an IOO perspective, it should be noted that even CAT applications that appear to be dependent solely on vehicle-to-vehicle (V2V) communications—merging onto a highway, for example—will depend on infrastructure information in the form of maps and controls. The (non-automated) vehicle-to-infrastructure (V2I) safety use cases will carry forward into a CAT context. As this transition occurs, the ADS, rather than a human driver, will consume and use the safety message. The initial priority V2I cases, like Red Light Violation Warning (RLVW) or Reduced Speed Zone Warning (RSZW), will in the automated case be providing notification of changing traffic control data, rather than the more general warning. Identification of potential cooperative automation applications is an ongoing process. The Automated Vehicles 4.0 3 (AV 4.0) document prepared by the National Science and Technology Council (NSTC) and the United States Department of Transportation (USDOT) identifies current research across transportation,

3 National Science and Technology Council (NSTC) and USDOT. Assuring American Leadership in Automated Vehicle Technologies: Automated Vehicles 4.0, https://www.transportation.gov/sites/dot.gov/files/docs/policy-initiatives/automated-vehicles/360956/ ensuringamericanleadershipav4.pdf, accessed January 21, 2020.

7 agricultural, military and other sectors. The USDOT AV 3.04 summary more specifically addresses potential AV applications on the nation’s roadways. Those applications could include for example vehicle platooning at reduced headway to increase capacity, speed harmonization to reduce potential congestion bottlenecks, cooperative lane change and merge to reduce flow disruption at interchanges, and signalized intersection approach and departure with signal phase and timing (SPaT) data. The initial application use cases being developed to demonstrate cooperative automation in the USDOT’s CARMASM program5 include merging onto highways, speed reduction for localized road weather conditions, automated path clearance for emergency vehicles, and navigating a signal-controlled, single-lane work zone.

IOO Guiding Principles for CAT Infrastructure

The Need and Basis for GPs

Most of the information and discussion around the advent of connected and automated vehicles has been focused on the operations and benefits of the vehicles themselves. The roadways and associated infrastructure have been a backdrop or, at best, enablers for the new vehicle technologies. The information currently available on these vehicles and the cooperative environment in which they operate tends to explore the applications, technologies, or stakeholder perspectives while deferring discussion of the infrastructure and system operations. In this context, IOOs have direct control of and responsibility for the transportation infrastructure and its interaction with users and vehicles. Vehicle automation affects those interactions in new systemic ways. Just as IOOs work now with human drivers to enhance safety and mobility with roadway design, traffic controls, and education, IOOs will have new opportunities to adapt those strategies to automated driving systems. IOOs need a more comprehensive framework for understanding and dealing with the systemic effects of automated vehicles being deployed across the transportation infrastructure. They need to understand the role of communications and data in enabling cooperative automation. They need to have a broader view of the CAT ecosystem and its development, deployment, and operations. They need to see the potential benefits and opportunities, as well as the risks and costs. The guiding principles for CAT infrastructure are intended to help bridge these gaps.

4 USDOT. Preparing for the Future of Transportation: Automated Vehicles 3.0. https://www.transportation.gov/sites/dot.gov/files/ docs/policy-initiatives/automated-vehicles/320711/preparing-future-transportation-automated-vehicle-30.pdf, accessed Janu- ary 21, 2020. 5 https://cms7.fhwa.dot.gov/research/research-programs/operations/carma-overview, accessed January 21, 2020.

8 Objectives of the GPs The objectives of the GPs are to: H Promote safety, mobility, equity, and operations efficiency in CAT H Facilitate collaboration H Support legislative and regulatory frameworks H Support interoperable deployments through standards, application development, deployment guidance, experience reporting H Educate the workforce H Provide context on CAT to better inform the public

GPs and Concepts

The GPs address five dimensions of IOO development of CAT. The functional core of CAT is Automation of vehicles and the Guiding Principles for CAT infrastructure that vehicles share with other CAT users. Data enables the automation operates. The vehicles and infrastructure may be the devices being automated, but data powers the automation. Communications enable the data interactions between the vehicles, infrastructure, and users. Operations capture and enact decisions about how the transportation system is automated. Collaboration creates an environment that values and incorporates the needs and objectives of all CAT participants. These five dimensions will work together to provide a seamless, consistent, and integrated CAT ecosystem across the nation. The guiding principles were formally adopted by the AASHTO Board of Directors at its 2019 Annual Meeting.

Automation

As described by the SAE model for levels of automation, vehicle automation is an ongoing incremental process. Vehicles already contain many automated GP 1—Automation functions that improve the safety and convenience of the driving tasks (for Support increased vehicle example, anti-lock brakes, adaptive cruise control, or parking assist). These automation to improve systems to date are localized to the vehicle and its independent operation. traveler safety, mobility, equity, and efficiency. Automation could proceed down a continuing path of autonomous operations, but this is inherently limiting. It leaves the vehicle operating according to its own sensors, to its own purposes. But vehicles share the infrastructure with other vehicles and users, many of which are significantly vulnerable to decisions that might be made inde- pendently by those autonomous systems. A more comprehensive view of the transportation system and its operations demands that automation be cooperative. CAT views transportation as a set of interactions between all the system users—passenger vehicles, commercial vehicles, transit, motorcycles, bicycles, micromobiles, pedestrians—supported by an infrastructure that facilitates safety, mobility, equity, and efficiency in their interactions. IOOs have a unique perspective and responsibility for automation of the transportation system as a whole.

9 As such, collaboration among stakeholders will be needed to establish frameworks for CAT that can address the needs and interests of all users. First steps GP 1A—Collaborate with toward a general framework have been undertaken by several stakeholder CAT stakeholders to define groups. The Federal government has provided its view of vehicle automation and establish national CAT and its system interfaces in its AV 2.0,6 3.0,7 and 4.08 visions. The American As- frameworks. sociation of Motor Vehicle Administrators (AAMVA) Autonomous Vehicle Infor- mation Sharing Group has developed several resources on AV, including guides for testing and deployment of autonomous vehicles9 and for testing drivers in vehicles with driver assistance systems.10 The National Association of Counties (NACo) provides and introduction and guidance in an on-line toolkit.11 A more overarching view is needed and is being addressed in and among IOOs and USDOT in efforts such as the AASHTO National Strategy for Highway Automation and the USDOT Highway Automation Concept of Operations. Participation of IOOs in defining and establishing this CAT framework is essential to assuring a system-wide infrastructure-based perspective. IOOs should seek to become involved in and support industry initiatives such as the CAT Coalition, and to participate with AASHTO, ITSA, ITE, and other associations in their respective CAT working groups. A “framework,” as such, represents a set of concepts for CAT. It needs a layer of mechanisms within the framework to provide tools and structures to be GP 1B—Support the development and used in development and deployment. Building these tools is an emergent implementation of standards, process, where CAT stakeholders begin to see where their needs and inter- regulations, training, and ests attach themselves to the framework. The tool kit will include, at a min- education efforts for CAT. imum, standards for component systems and their interactions, regulations for development and operations, training, and education. This is a far-reaching effort with implications for fundamental transportation standards and resources. AASHTO’s A Policy on Geometric Design of Highways and Streets (the “Green Book”)12 for example may need in future editions to consider the provisions for dedicated AV facilities, or for mixed human-driven and automated vehicle traffic. Previous revisions of the Manual on Uniform Traffic Control Devices (MUTCD)13 have made impressive provisions for non-vehicular and vulnerable users, and may in the future need to further consider the emerging aspects of automated vehicles. state and local governments will need to review and potentially revise their engineering standards, specifications, and vehicle codes for automated vehicles and CAT.

6 NHTSA. Automated Driving Systems (ADS): A Vision for Safety 2.0. https://www.transportation.gov/av/2.0, accessed January 21, 2020. 7 USDOT. Preparing for the Future of Transportation: Automated Vehicles 3.0. https://www.transportation.gov/sites/dot.gov/files/ docs/policy-initiatives/automated-vehicles/320711/preparing-future-transportation-automated-vehicle-30.pdf, accessed Janu- ary 21, 2020. 8 National Science and Technology Council (NSTC) and USDOT. Assuring American Leadership in Automated Vehicle Technologies: Automated Vehicles 4.0. https://www.transportation.gov/sites/dot.gov/files/docs/policy-initiatives/automated-vehicles/360956/ ensuringamericanleadershipav4.pdf, accessed January 21, 2020. 9 AAMVA. Jurisdictional Guidelines for the Safe Testing and Deployment of Highly Automated Vehicles, May 2018. 10 AAMVA. Guidelines for Testing Drivers in Vehicles with Advanced Driver Assistance Systems, August 2019. 11 https://www.naco.org/resources/featured/connected-autonomous-vehicles-toolkit, accessed November 30, 2019. 12 AASHTO. Policy on Geometric Design of Highways and Streets, 7th Edition. 13 FHWA. Manual on Uniform Traffic Control Devices, 2009 Edition.

10 IOOs should work together, with other CAT stakeholders, with their related state and local agencies, and within their organizations to develop these tools. It will be to the IOOs advantage to monitor and participate in CAT standards development, support development of regulations within their states and local governments, and adapt workforce development resources to address CAT. Agencies recognize that infrastructure deployments are by definition local. As the tools for working within a CAT context become available, IOOs will GP 1C—Plan and execute need strategies for planning, developing, deploying, and operating an infra- readiness strategies for structure supporting CAT and its users. Strategies will need to leave oppor- CAT planning, development, tunities for stakeholder involvement in localization to community needs and infrastructure deployment, and objectives. Many states and local governments have started this process with operations. strategies for engaging with AV development and deployment. This will need to be expanded to a full consideration of the agencies’ own automation efforts and engagement with CAT stakeholders. This will be a long process that evolves with the increasing levels of automation and more widespread deployments. Lessons learned in the development and deployment of V2I capabilities will be valuable in both planning and execution of CAT capabilities. IOOs can and should be planning for CAT readiness, even as the frameworks and tools are developing. They can include CAT concepts in long-range planning. They can look for opportunities to engage in prototype CAT projects to build awareness, understanding, and experience. And they can pursue AV and CAT funding opportunities with USDOT and with private development partners.

From a practical perspective, CAT infrastructure developments build on GP 1D—Deploy or accommodate prior generations of ITS and connected vehicle technologies, systems, and communications infrastructure applications. In particular, the cooperative aspects of CAT involve two-way to support CAT data and message exchange between vehicles, users, and infrastructure, just as in messaging among vehicles, connected vehicle applications. This will require data and communications users, and infrastructure infrastructure among all modes and at all levels of vehicle automation. components. IOOs can accelerate CAT-readiness by looking for opportunities to leverage and expand their ongoing ITS and CV deployments, especially in communications. They can acquire and/ or deploy very fast, high-bandwidth backhaul communications networks that are extensible and expandable. They can build CAT perspectives with respect to siting, bandwidth, and cybersecurity into planned communications deployments. They can look for opportunities to leverage communications investments with other public agencies and private partners. CAT application needs and objectives can also be used to inform early infrastructure developments. The initial AV applications are growing out of GP 1E—Collaborate with OEMs autonomous and V2V applications. Infrastructure requirements for these and other CAT stakeholders to applications may be more related to uniformity and maintenance, and less to identify priority applications technological complexity. At a minimum, AV applications want to have clear, and infrastructure development. consistent, and well-maintained pavement markings and signage. Applications over the longer term will grow to reflect fully cooperative TSMO strategies. IOOs will have increasing opportunities to affect system performance in real time using dynamic traffic controls. Existing variable speed limit zones, for example, could expand to become real-time dynamic limits along entire corridors or across regions to mitigate flow breakdown and preserve mobility in congested conditions.

11 Data

The cooperative aspect of CAT requires reliable secure data exchanges between vehicles and the transportation infrastructure. These data ex- GP 2—Data changes will also require cooperation across and among CAT stakehold- Achieve a connected vehicle ers to assure seamless data exchange capabilities, standards, security, ecosystem that enables reliable, permissions, deployment, and operations. secure V2I data exchanges in order to support cooperative automated For IOOs, this means planning for de- transportation to improve traveler ployment of an information infrastruc- GP 2A—Make plans for safety, mobility, equity, and ture to support automation. This will connected infrastructure efficiency. be similar to and an extension of the deployments supporting V2I work already undertaken for ITS and applications available to CAT for CV, with this key distinction: data stakeholders. for vehicle automation will ultimately be intended for and consumed by ADS rather than human drivers. Although artificial intelligence and machine learning methods are very powerful, they need high quality data on which to operate. Data will generally need to be of consistent high quality, availability, and timeliness from reliable sources. Data will also need to be precise and detailed to support ADS operations. Driving regulations and traffic controls applicable within an agency’s jurisdiction will need to be codified in a form accessible to ADS. Work zone data sets, for example, may need to describe the location of drivable lanes within a few centimeters of precision. Agencies can begin working toward a CAT ecosystem by planning for the digital infrastructure that enables cooperation. They can review and amend data policies and procedures for the exchange of data between vehicles and agency systems. As described for GP 1D, they can plan for roadside and backhaul communications coverage consistent with application deployment, growing over time to complete coverage of their transportation networks. They can work with agency IT support to plan for developing and acquiring computing infrastructure for CAT applications. And they can review and plan for transportation systems and operations data to be made available to CAT applications. The ability to cooperate implies that communications are mutually under- stood. Cooperation depends on development of and commitment to stan- GP 2B—Build consensus dards for communications and data exchanges among the stakeholders, and among CAT stakeholders on on consistent interpretation and implementation of those standards. Prior standards and guidance for development of standards for ITS and for CV applications has provided a solid V2I communications and data exchanges. foundation for further CAT developments. In particular, the SAE standards developed for DSRC V2V and V2I messaging describe many of the fundamental exchanges needed. Standards will nonetheless need to be extended and supplemented to support automation across the CAT ecosystem. ADS, for example, use high-definition maps working with their sensors to plan routes and maneuvers. Current standards for CV applications are not explicitly designed for high-definition maps and messages, and may need revision or extension to fully enable AV-ready applications. The current human-directed traveler information messages based on the Traffic Management Data Dictionary (TMDD)14 will similarly need to become more precise for ADS interactions in their description of incidents, work zones, and other events.

14 ITE. https://www.ite.org/technical-resources/standards/tmdd/, accessed January 27, 2020.

12 Standards furthermore need consensus among their users to be effective. Perspectives of data providers and users are needed to assure that data definitions, quality standards, latencies, and other performance requirements are useful and achievable. IOOs can and should seek to be involved in standards development, whether as originators of data about the infrastructure and its operations, or as users of data from vehicles. A complex ecosystem with numerous points of interaction, and risks to life and property needs a high level of trust in the information being exchanged. GP 2C—Coordinate with CAT IOOs generally have experience with information security challenges as part stakeholders to establish of their ITS infrastructure deployments. Developing a trusted CAT ecosys- a nationwide approach for tem, however, introduces new sets of attack vectors both at the edges of and securing data exchanges within the networks, across multiple communications media. It is essential between vehicles and the that the communications between vehicles and infrastructure demonstrate infrastructure. secure and reliable data exchange. To that end, all stakeholders in the CAT data supply chains will need to work together to establish security performance requirement and standards to build that trusted ecosystem. IOOs can plan for secure data exchange by first looking within their own systems and operations to provide a high level of security and reliability. Data security at the infrastructure interface to vehicle systems needs high assurance of quality and reliability as a basis for preserving quality and reliability in the exchange. Infrastruc- ture virtual points of presence will then also need to be secured as in any other network. CAT systems and applications may bring new complexities, but will depend on strategies and plans similar to and expanding those that would already have been put into place for ITS networks. Just as in those ITS applications, IOOs will need to work with the network and communications system providers and with the data exchange partners (vehicle system and ADS developers) to build end-to-end secure data exchanges. Having agreed on standards and provided for security, the CAT ecosystem is prepared for the actual exchange of information between vehicles and GP 2D—Utilize all existing and infrastructure. This is however not just an exchange of data, but a coopera- future V2I standards in order tive exchange of value. ADS-equipped vehicles need information from IOOs to achieve two-way data exchanges between connected about roadway configurations, speed limits, the status of signal controls, infrastructure and production lane realignments around work zones and incidents, and hazardous roadway OEM vehicles. conditions. In exchange, IOOs can receive data from vehicles about intended routes, current paths, and sensed local conditions. The value to the ADS- equipped vehicle is being able to reduce travel time and the risk of conflicts. The value to IOOs is in aggregating the data to get views of network conditions and in being able to manage traffic controls to reduce safety risks and preserve mobility. Adding automation increases the potential value of V2I data exchanges and provides incentives for deployment that are less obvious in a merely connected ecosystem. Human operators may not choose or be able to act on information outside their experience or context. The volume of and speed with which data will be available are likely to exceed a typical human driver’s capacity to take in the data. Automation will increase the reliability, accuracy, and efficacy of applications that would otherwise depend on human interaction and response. The perception and reality of return on infrastructure investment should be higher and more direct than without the automated components.

13 IOOs should be investigating the opportunities for using data provided by vehicles as a means to improving their own performance measures through automation. For example, how might access to individual and aggre- gated vehicle trajectories provide faster awareness and response to incidents? What mobility improvements might be available if speed harmonization and smoothing could mitigate flow disruptions before flow breakdown? Could higher resolution road condition awareness from data reported by vehicles be used to better respond to weather events, improving safety while making more efficient use of agency resources? While value can be demonstrated by enabling data exchange for automation in particular locations and vehicles, societal value is unlocked only when that GP 2E—Work together to develop grows into a national deployment. Manufacturers and developers want to and communicate a national know that the data will be reliably available across ODDs. IOOs benefit when roadmap for deployment of a costs of development are shared across a larger set of deployers. Individuals secure and standards-compliant purchasing vehicles and using automated services become more confident connected infrastructure. in the technologies and applications when they see widespread access and acceptance among other users. A national roadmap, even if not prescriptive, provides a unifying vision and template for deployment and public adoption. The Eisenhower Administration’s vision for the Interstate Highway System engaged the public and industry in such a way that the initial demonstrations grew to a national network over 40 years. Roughly the same timeframe will be needed from initial deployments to a mature CAT ecosystem, considering both the extent of the infrastructure needs and turnover of the vehicle fleet. A similar vision may be needed to initiate the time- line. It is important that the communications and roadmap furthermore emphasize the security and standards compliance of the connected infrastructure deployment. The safety of the infrastructure, communications, and automation applications require that they be secured from misbehaviors and bad actors, and made resilient to natural and external system events. Infrastructure deployments will need to adhere to consistent implementa- tions of standards. IOOs can jumpstart this process by participating in national and regional dialogs on roadway automation and CAT. There are ample opportunities to participate in broad industry interest groups and conferences working on cooperative automation and standards. Agencies can leverage progress made in regional cooperation on ITS and connected vehicle applications through corridor coalitions, AASHTO regions, and technical committees toward CAT readiness. They can be aware of and participate when appropriate in teams and focus groups working to develop CAT research and deployment planning.

Opportunities for and benefits of systemic automation will emerge from the GP 2F—Work together with OEMs CAT ecosystem as connected infrastructure, vehicle deployments, and data and mobility service providers exchanges expand. A maturing CAT ecosystem will, for example, extend to ensure seamless operations, individual awareness of traffic conditions so as to both reduce imminent risk mindful of the appropriate and enhance mobility for optimal travel times. It will expand traffic, user, and mobility strategies, based on service provider data models for Mobility on Demand (MOD) and Mobility as two-way data exchange. a Service (MaaS). It will enable fleet operators—whether in freight, transit, emergency, or maintenance services—to better preposition and respond to user demand and external events. It will enable IOOs to respond to network conditions with more specific and timely operations strategies to preserve safety and mobility. The interests of stakeholders in a fully connected CAT ecosystem will work together to enhance and reinforce the safety, mobility, equity, and efficiency of system operations as a whole.

14 Telecommunications

Telecommunications are the essential means of cooperation in the CAT GP 3—Telecommunications ecosystem. The Federal Communications Commission (FCC) allocated Protect and utilize the 5.9 spectrum in the 5.9 Gigahertz (GHz) band (the “Safety Band”) in 1999 for Gigahertz (GHz) spectrum ITS applications enhancing safety and mobility. That allocation specified designated for “operations related using DSRC technology in the 5.9 GHz band to assure interoperability to the improvement of traffic among devices and applications. flow, traffic safety and other intelligent transportation service Societal demand for communications technology has expanded signifi- applications.” 15 cantly since then, creating competitive interest in the spectrum allocated to the transportation Safety Band. As such, telecommunications technology providers are pursuing new technologies and opportunities to further leverage the 5.9 GHz band for transportation safety and other applications. The FCC is reviewing these technologies and applications and revisiting the dedicated allocation. IOOs can protect the public interests first by using the 5.9 GHz spectrum for its intended applications, and then also by monitoring developments with the FCC and defending its allocation exclusively for transportation safety. Some of the interest in other uses and technologies for the 5.9 GHz band is based on a perception that the pace of deployment has not been sufficient GP 3A—Continue to use the 5.9 to merit continued exclusive use for DSRC-based transportation safety and GHz spectrum in deploying connected infrastructure mobility applications. In practice, the rate of deployments has accelerated systems including but not tremendously over the last ten years. The 2012 Ann Arbor Safety Pilot Model limited to the FCC-licensed Deployment “utilized connected vehicle technology in over 2,800 vehicles Dedicated Short-Range and at 29 infrastructure sites at a total cost of over $50 million dollars in order Communications (DSRC) to test the effectiveness of the connected vehicle technology.”1615The USDOT technology. Connected Vehicle Pilot Deployment Program1716 sponsored extensive pilot projects in three states with thousands of equipped vehicles and dozens of roadside units. The CAT Coalition’s SPaT Challenge1817 has championed deployments of infrastructure for criti- cal safety and mobility messaging applications such as SPaT for RLVW and to improve first response and transit schedule adherence. The Coalition is also challenging IOOs to deploy in-vehicle technology in the Connected Fleet Challenge.1918 IOOs can and should continue to deploy applications and demonstrate the value of the 5.9 GHz spectrum for transportation safety and mobility. Applications using the proven and FCC-licensed DSRC technology are supported by 20 years of USDOT and state research and practical experience across the IOO community. More recent deployments are demonstrating application in the 5.9 GHz band based on C-V2X technology for similar

15 Amendment of Parts 2 and 90 of the Commission’s Rules to Allocate the 5.850-5.925 GHz Band to the Mobile Service for Dedicated Short Range Communications of Intelligent Transportation Services, ET Docket No. 98-95, Report and Order, 14 FCC Rcd 18221 (1999) (DSRC Report and Order). 16 USDOT. “Connected Vehicle—Safety ITS Benefits, Costs, and Lessons Learned: 2018 Update Report,” accessed January 3, 2020 at https://www.itsknowledgeresources.its.dot.gov/its/bcllupdate/pdf/BCLL%20CV%20Safety%202018%20v2.pdf. 17 https://www.its.dot.gov/pilots/, accessed January 27, 2020. 18 https://transportationops.org/spatchallenge, accessed January 3, 2020. 19 https://transportationops.org/connected-fleet-challenge, accessed January 3, 2020.

15 applications. Deploying connected applications in spite of technological uncertainty sends a message that transportation safety and mobility continue to warrant a national dedication of communications spectrum for those purposes. Given that the Safety Band is secured for transportation applications, it is important to continue to invest in technologies providing the best return on GP 3B—Encourage the that investment. Each new such technology needs to be subjected to rigorous introduction, rigorous testing, testing and evaluation. Each such technology will have to support interop- and evaluation of new 5.9 GHz erability and compatibility with the existing applications and comply with communications technologies that support interoperability the relevant performance standards in order to protect past investments of and that comply with both public and private capital. Encouraging technological innovation should performance standards. assure a focus on the objective—enhanced safety, mobility, equity, and efficiency—rather than on the particular means of implementation. IOOs are demonstrating commitment to ongoing innovation, for example, in testing and evaluation of DSRC and C-V2X in similar applications. This approach should be supported and extended by the IOO community as even newer technologies are proposed and introduced. Any such new technologies are nonetheless subject to the established FCC GP 3C— Work with new 5.9 technical and service rules. IOOs are unlikely to pursue or develop new com- GHz technologies within the munications technologies within their own organizations, but are frequently established FCC technical and approached by developers with new concepts. Any such engagement should service rules. work within the existing rules, or be subject to appropriate license exception. It would not be appropriate to pursue or deploy solutions that intentionally operate outside of or counter to the intent of rules and guidance. Development and deployment of Safety Band technology is, of course, a GP 3D— Continue coordination mutual commitment of IOOs and automotive OEMs. AASHTO has worked with OEMs to support long-term with OEM representatives on connected vehicle initiatives for over fifteen investment decisions regarding years to plan for effective utilization of the 5.9 GHz spectrum and deploy- the deployment of communications technologies utilizing the 5.9 GHz ment of connected vehicle applications. That effort should continue. A spectrum. mutual understanding of the objectives, opportunities, and challenges will go a long way toward visionary and rational long-term investments. A further clarifying point: technological innovation must acknowledge the need to assure that applications remain interoperable across US and North GP 3E— Work with new 5.9 GHz American jurisdictions and OEM fleets. The very premise of CAT and con- technologies within the established nected vehicles is that vehicles be able to communicate with each other FCC technical and service rules. and the infrastructure as they move throughout the road network. Social and economic ties necessitate that such vehicles and communications be able to operate seamlessly across the continent.

16 Operations

From an IOO perspective, automation, data systems, and telecommuni- GP 4—Operations cations are only the means to the development of operational strategies Develop CAT strategies that that improve traveler safety, mobility, equity, and efficiency in a CAT enhance existing transportation ecosystem. Facilitating vehicle automation with standardized messag- system operational capabilities to ing and communications is a valuable technical and economic objective, improve traveler safety, mobility, but it falls well short of the potential of a fully realized CAT ecosystem. equity, and efficiency. Individual vehicle safety and mobility are important, but IOOs will have an opportunity with CAT to implement strategies with network-wide effects. Connectivity and automation applied to operations will improve safety and relieve traffic congestion. Imagine traffic operations optimized to reduce the risk of crashes before they happen, or to mitigate flow disruptions and breakdown. Connectivity and automation may not result in zero deaths and zero congestion, but those goals are not achievable without them. CAT strategies are likely in many cases to be extensions of traditional and ITS solutions to longstanding transportation challenges, enhanced by GP 4A—Identify operational expansive data analytics. For example, deployment of traffic controls as capabilities that can be leveraged to improve the markings, signs, and signals to preserve safety could become a set of just- effectiveness and/or cost in-time traffic control messages. Or demand management strategies might effectiveness of existing and use dynamic routing, access control, and metering. Capacity enhancement future operational investments. might take the form of reduced headway for AVs rather than additional lanes. Traffic smoothing using demand-aware speed control might help avoid flow breakdowns entirely. All of these strategies envision leveraging the ability to communicate dynamically with vehicles, and the higher degrees of compliance implicit with algorithmic data-driven vehicles, to capture the margin currently used in accommodating variable skill and awareness among human drivers. IOOs can and should investigate appropriate opportunities to improve transportation system performance through automation. CAT is not just about supporting vehicle automation through provision of maps and safety messages. IOOs should be looking at network-centric strategies to improve system performance as well as at the needs of vehicles operating on the network. Significant and complementary national efforts toward articulating CAT GP 4B—Adopt the outcomes strategies have been convened and will be continuing to build frameworks of national efforts to develop for CAT. USDOT convened and sponsored workshops in a National Dialog on new strategic operational Highway Automation in 2018. Summaries of the workshop dialogues provide deployment initiatives that 20 expansive perspectives on automation and cooperative strategies.19 A Na- utilize emerging connected, tional Strategy for Highway Automation (NHSA) is being developed by automated vehicle technologies, AASHTO’s Committee on Transportation System Operations (CTSO) Highway and mobility services. Automation Task Force working with contributions from IOOs, USDOT, and trade associations.2120 USDOT will be building on the success of the National Dialog with development of a Highway Automation Concept of Operations. While each effort has its own ar- eas of focus and perspectives, they share similar objectives, in that they all envision a cooperative automated future.

20 https://ops.fhwa.dot.gov/automationdialogue/, accessed January 21, 2020. 21 https://systemoperations.transportation.org/wp-content/uploads/sites/22/2019/09/National-Strategy-on-Highway-Automa- tion-CTSO-Update-8_26_19.pdf, accessed January 21, 2020

17 for transportation on the nation’s roadways. They have been and will continue to be informed by the collective experience with connected vehicles, state and local development and deployment of CV and AV projects, smart city initiatives, and other programs like the Advanced Transportation and Congestion Management Technology Deployment (ATCMTD) and ADS Demonstration grants. IOOs can leverage these efforts initially by being aware of and participating wherever possible in their pro- ceedings. As they continue to develop, IOOs can consider application of the strategies to their particular agency circumstances and opportunities, eventually building them into planning for CAT development and deployment. New strategies, models, and technologies present risks as well as opportunities. The challenge is to balance the potential upsides with reasoned GP 4C—Identify any regulatory barriers to enable the safe attention to risk. This may require review of and consideration for regulatory deployment of new capabilities barriers that might unnecessarily hinder innovation while protecting the pub- as a result of emerging lic interests. The automation of driving tasks, for example, changes the deci- technologies and mobility sion maker and actor from a human driver to algorithms and control systems. service models. State vehicle and traffic codes are written to license drivers and vehicles, but AVs are a hybrid of both. Capturing the benefits of automation for operations strategies may require changes to certain rules within those codes. IOOs can and should work with their related government agencies and legislative bodies to identify and evalu- ate existing codes. AASHTO, AAMVA, and other agency associations can serve as collectors and repositories of other agencies’ perspectives, experiences, and findings, facilitating the process. The potential benefits of operations in a CAT ecosystem will drive expansive GP 4D—Embrace CAT strategy innovation as the technologies develop. Entrepreneurs and early-adopting interoperability and uniformity. agencies will create virtuous circles of new technologies feeding development of new strategies to address particular ODDs and operational issues. Over time, these will likely converge into recognized standard products, services, and strategies. As they develop, however, it will be in all parties’ best interest to work to a common understanding of the interfaces between the components. Interoperability and uniformity among strategies will facilitate development, deployment, and user acceptance. IOOs have used this pattern extensively in their V2I application deployments working with DSRC providers and with each other through the Deployment Coalition. A speculative example for CAT applications: automated variable speed limits (VSL) might be used to assure continuity of flow under congested and variable weather conditions. The VSL strategies would need to operate similarly across jurisdictions to assure that developers have a common set of expectations as vehicles move from one city or state to another. This consistency in data sets and interfaces could also contribute to revenue opportunities with tolling, congestion pricing, and mileage-based user fees. IOOs will at some level be partners in CAT strategy development even if driven by commercial technology and service innovation. They may actively sponsor or passively enable those developments on their roadways, but should in any case assure interoperability of the resulting applications with their existing systems and communications, and monitor system performance to assure agency standards are maintained. They can become more integral parts of strategy development by participating in interagency and industry groups working on concepts and standards. As the core practitioners for TSMO, IOOs have the best understanding of the opportunities and challenges that might emerge from development of the CAT ecosystem.

18 Collaboration

The cooperative emphasis within CAT makes it essential that the tech- GP 5—Collaboration nology and application developers and deployers collaborate in planning, testing, and demonstration of applications. All parties to develop- Collaborate and communicate ment and deployment will have their own particular interests, but a fully with OEMs and mobility service providers in the planning, testing, interoperable CAT ecosystem will need them to acknowledge common and demonstrations of CAT interests and the overarching safety, mobility, equity, and efficiency applications to support eventual goals for CAT. interoperability and to achieve The IOO community as represented positive impacts on safety, by AASHTO, ITE, and ITSA has set up GP 5A—Encourage OEMs and mobility, equity, and efficiency and been working through CAV policy, mobility service providers technology, communications, and to interact with established operations committees for many years. The automotive OEMs, individually IOO associations’ existing and evolving committees and and through groups such as the Collision Avoidance Metrics Partnership working group structures (CAMP), have been frequent and valuable discussion partners. Mobility when pursuing feedback and/ service providers have become more involved as their marketplace presence or involvement regarding has grown and business models are refined. applications intended for These interactions need to continue. The commercial partners are looking positive impacts to the for IOO input as a means of assuring that their development plans will be infrastructure system. congruent with IOO intentions. IOOs can benefit from hearing perspectives from vehicle, technology, and service developers. Using IOO and trade associations as forums for discussion provides breadth of engagement and exposure. These engagements can be made even more valuable when IOO representatives are actively engaged in association committees and working groups. AASHTO, ITE, and ITSA each have their respective committee structures and have jointly convened the CAT Coalition as a way of pooling interests and perspectives to provide non-competitive input to OEMs, mobility GP 5B—Support activities within service providers, and other commercial interests. IOO associations’ committees and IOO management can further these opportunities by identifying individu- working groups to collaborate on creating non-competitive input als and enabling participation from agency staff within their administrative to OEMs and mobility service and budget constraints. Participation then becomes a two-way conduit for providers on CAT-related topics. providing each agency’s perspectives while bringing back news and opportunities for further influence. Opportunities for public agency and commercial party engagement may GP 5C—Be open to individual further extend to collaboration on technology development and applica- IOOs (or groups of IOOs) collaborating competitively with tion demonstrations. Agency coalitions and pooled funds have previously one or more OEMs in situations provided these types of engagements in ITS and connected vehicle devel- where innovation, exploration opment and might serve as models for emerging CAV opportunities. These of products, or funding selective contractual arrangements have the advantage of accelerating opportunities benefit from such development by enabling the commercial parties to negotiate terms that collaboration. protect their commercial interests while advancing the IOOs’ broader goals.

19 The results of these competitive efforts are nonetheless valuable to the IOO community as a whole. Wherever possible, IOOs can advance the develop- GP 5D—IOOs participating in ment of an interoperable CAT ecosystem by sharing their experiences and competitive collaboration outcomes with other agencies. IOO association committees, work groups, and with OEMs shall seek to share other meetings provide opportunities for presentations, focus groups, and outcomes of the activities, as allowed by their agreements, panels that can communicate the findings in an open forum while reserving to encourage and support a control over the depth and particulars of disclosure. Other IOO parties are collaborative CAT environment. then encouraged and informed, moving all parties closer to the CAT goals of improving safety, mobility, equity, and efficiency.

Applying the CAT Infrastructure GPs

The development of the CAT ecosystem is in its early phases. In the United States, this will be a distinctly emergent process based on cooperation among a diverse set of public agencies, commercial developers, and academia. It will not be developing under the presumption of a top-down plan. As noted in AV 4.0,

The U.S. Government will be proactive about AVs and will provide guidance, best practices, conduct research and pilot programs, and other assistance to help stakeholders plan and make the investments needed for a dynamic and flexible future for all Americans. We will also prepare for complementary technologies that enhance the benefits of AVs, such as communications between vehicles and the surrounding environment, but will not assume universal implementation of any particular approach.—AV 4.0

As such, the Guiding Principles have implicitly assumed that IOOs have common interests in CAT development, but may nonetheless differ in the specifics of their deployment processes. The GP applications described here are intended as starting points for further consideration. AASHTO, ITE, and ITSA, on behalf of the IOOs, will monitor and initiate new activities as the CAT ecosystem develops.

CAT and IOO Processes

Although the eventual operational capabilities of a CAT ecosystem may change the way vehicles and infrastructure work together, the deployment of infrastructure in support of CAT has to work through familiar IOO processes. This reality implies two gaps in perspective that may challenge an IOO’s ability to plan for CAT. First, it is not yet clear in what circumstances vehicles and infrastructure will need to operate cooperatively. This makes it difficult to foresee where to prioritize infrastructure investment for CAT. Second, new technology development and typical infrastructure deployment move at significantly different paces. New AV technological concepts are being announced and redirected on cycles as short as a few months. Infrastructure projects at State DOTs may be in planning and development for 20 years, even when the technology is up to date as it is being deployed. The perceived gaps may however be an issue of perspective more than reality. Technological change appears to move quickly because it is focused on how things are done. Infrastructure appears to need a longer view because it is focused on what needs to be done, according to strict priorities. When the technology is viewed for what it might do (its use cases) and infrastructure for how it contributes to what needs to be done (its use cases), the two can come together.

20 As a practical example, AVs intended to operate on public roadways today have to obey traffic signals just as non-AVs do. No infrastructure change needs to be made for that use case. There are advantages, however, in providing SPaT data to the AVs so that they can anticipate signal operations (another use case) and operate where visual sensors may not be able to see the signals (another use case). A far-off cooperative future with only AVs, or the ability to provide human drivers with other ways of safely navigating intersections, might re- move the physical signals (yet another use case). Planning for the signal infrastructure gets broken down into a stream of incremental changes rather than a single question of how to deploy signal systems that support AVs. This perspective suggests that IOOs may benefit most from looking at specific infrastructure projects that support AV deployment use cases, rather than investing too heavily in top-down infrastructure plans. The broad high-level view looks further into the future, but still can only deploy for specific use cases and technologies. This is precisely the approach that worked for the first connected V2I deployments, and before that for suc- cessful ITS deployments. Each phase of an IOO’s development and deployment processes may then need to use those views in looking at infrastructure supporting CAT. The context for CAT infrastructure deployments is played out in industry and regulatory initiatives like those that were instrumental in CV and ITS deployments. AASHTO, ITE, ITSA, and other associations, USDOT and its constituent agencies, and TRB with the research community can facilitate monitoring and reviewing those high-level movements. IOOs participating in the CAT Coalition both support and have direct access to these processes in both executive sessions and technical working groups. Their activities include: H Offering platforms to discuss and integrate CAT needs of local transportation agencies in consideration of policy developing initiatives with ongoing AV efforts; H Collecting and documenting member viewpoints on potential future policies, rulemakings, cooperative agreements, or guidance that may affect CAT efforts; H Collecting and documenting IOO experiences with and viewpoints on CAT efforts and projects (for example, perspectives on Na- tional Highway Transportation Safety Administration (NHTSA) rule-making, security systems, and telecommunications); H Offering insights into high-level decision-making processes within the industry as related to technology and communications; H Assessing the potential benefits and risks associated with CAT deployment; H Providing strategic deployment guidance (like this document); H Recommending policies and national CAT deployment approaches to association members; and H Providing input to technology developers, vehicle manufacturers, and the private sector about how they can work with IOOs on CAT deployments. Within the larger CAT context, individual agencies can use those association and Federal initiatives to inform their own planning processes. Project identification will depend on factors much like any ITS deployment, from operational needs to available agency funding and grant opportunities. Partnerships with technology developers and vehicle manufacturers, and association with other agencies may inspire or kickstart state and local efforts. Aspects of CAT deployment projects supporting interoperability, security, and future research will likely follow the developing industry standards. The objective and scope of particular projects, on the other hand, will derive from the agency’s own strategic research and operational needs. The particulars of IOO CAT infrastructure application development, deployment, testing, and operations are likely to work out similarly to previous generations of CV and ITS projects. The vision for the CAT ecosystem is

21 still developing, and it would be premature to identify particular issues and guidance on those processes. It is clear, however, that any of these activities with CAT will depend more directly on partnerships and cooperation with the automation systems technology providers and vehicle manufacturers.

Preparing for CAT Infrastructure

Enough is known about the emerging CAT ecosystem to provide some guidance on preparing for CAT infrastructure for AVs even without specific use cases. These activities might be integrated into an agency’s ongoing system development, maintenance, and operations to the extent they can be supported by available agency resources. Agencies can and should stay connected to ongoing CAT development and deployment initiatives. Participating in the CAT Coalition, for example, extends an awareness of emerging issues, opportunities, and developments within the industry. Agencies can take an active role in issues that directly affect CAT infrastructure, like pres- ervation of the 5.9 GHz band for transportation safety. CAT applications from an infrastructure perspective are built on a foundation of prior TSMO strategies and applications in CV and ITS. Industry participation, training, workforce development, and capability building in TSMO will create advantages for CAT projects at all levels of cooperation and automation. Much of the initial vehicle automation efforts presume that traditional infrastructure features for traffic control will continue to be available for AV navigation and maneuvers. Improving and maintaining the state of repair of signage and pavement markings will facilitate AV development and the expansion of ODDs. Upgrading traffic signal and ITS equipment to the latest standards will enable CAT applications to build off their data and interfaces without project-specific additions or replacements. These upgrades should include deployment of fiber backhaul where applicable. Developing a CAT ecosystem will greatly expand the need for high-quality systems configuration and operations data. Agencies can start laying down a CAT information infrastructure by building and interfacing roadway design, asset management, signage, operations, and control data bases. Although OEMs, ADS developers, and map service providers are working on high-resolution maps for their own purposes, agencies may also consider upgrading their road network maps to centimeter-level precision with lane-level attributes (for example, width and pavement type), height restrictions, lane markings, and signage location data. Network and operations data exchanges with other agencies can anticipate similar opportunities with information services that will interact directly with AVs and fleets. The cooperative aspects of a CAT ecosystem demand reliable and consistent communications between vehicles and infrastructure. Agencies can facilitate their own operational needs and anticipate this demand by deploying digital network communications capabilities for I2I and I2V as far across the road network as is achievable.

22 Future Efforts

The Guiding Principles and Supporting Technical Concepts have been developed by a joint three-party AASHTO/ITE/ITSA Joint Task Force. Beginning in early 2019, a series of meetings resulted in the drafting of the core principles and reviews by committees and working groups within the three associations, from which the guiding principles were revised in response to the comments received. The final guiding principles were approved and adopted by AASHTO at its Annual Meeting in October 2019. This Supporting Technical Concepts document has been developed in recognizing that agencies, their industry partners, and associations would benefit from additional context and application description to supplement the guiding principles themselves. It is not intended to be a “final” or “complete” description of technical resources that will be needed to support CAT deployments. It does provide a starting point for what will be a continuous process of assessing and responding to evolving institutional goals, operational objectives, and technological developments for CAT. Priority initiatives and topics building on the Guiding Principles in the near term could include: H Expanding partner engagement to include CAT stakeholders from among other public agencies, vehicle manufacturers, ADS developers, technology providers, mobility service providers, and user groups; H Cooperation among public and private sectors in data governance and sharing, including identification of needs, privacy, and security; H AV, ADS, and infrastructure testing and evaluation; and H Planning and developing institutional, human, and financial resources toward deployment readiness. The Joint Task Force will be monitoring developments and amending these CAT Technical Concepts as appropriate, likely evolving into more formal technical support and exchange efforts.

23 Resources and References

All links provided in this section were accessed and current as of January 24, 2020.

CAT Coalition Activities and Resources

The CAT Coalition website is the essential first point of reference for information for IOOs related to CAT. It provides summaries of and links to CAT Coalition Working Group activities and resources, particularly for IOO experience in developing and deploying infrastructure in support of V2X communications and messaging. https://transportationops.org/CATCoalition/

National Cooperative Highway Research Program (NCHRP) Studies Related to CAT

The NCHRP, funded by the state departments of transportation and managed by the Transportation Research Board, started work in December 2014 under Project 20-102 to investigate CV and AV issues pertinent to CAT. The effort is being coordinated with the USDOT, private sector, and other efforts. The program summary document provides links to both completed project reports and descriptions of work in progress. http://onlinepubs.trb.org/onlinepubs/nchrp/docs/NCHRP20-102_CV-AV-Summary.pdf

USDOT/FHWA AV Resources and Projects

The USDOT provides a consolidated gateway to their AV programs and resources from which to navigate to program and project summaries. https://www.transportation.gov/AV The USDOT AV 4.0 activities, developed by USDOT working with the White House Office of Science and Tech- nology Policy, are a recent addition to the site. The AV 4.0 document contains an exhaustive list of Federal AV programs in its Appendix A. https://www.transportation.gov/av/4 USDOT’s prior AV program summary, AV 3.0, is focused on efforts within USDOT and its modal agencies. Ap- pendix C of the AV 3.0 document includes extensive lists of relevant AV/CV standards and standards development activities as of its 2019 date of publication. https://www.transportation.gov/av/3 The 2018 USDOT AV summary, AV 2.0, was developed by NHTSA to offer “a non-regulatory approach to automated vehicle technology safety”. While Section 1: Voluntary Guidance discusses AV concepts primarily directed to developers, Section 2: Technical Assistance to States provides foundational perspective on the role of States and IOOs in planning for AVs and CAT. https://www.transportation.gov/av/2.0

24 The USDOT in 2019 awarded $60 million in ADS Demonstration Grants to assess the deployment of AVs on the nation’s roadways. The resulting projects are generally underway and will provide perspective on the diverse opportunities and challenges of deployment. https://www.transportation.gov/av/grants The USDOT Data for Automated Vehicle Integration (DAVI) initiative is assessing the needs for and supporting data exchange for AV development and deployment. https://www.transportation.gov/av/data The Work Zone Data Exchange (WZDx) project is a foundational effort in facilitating and establishing voluntary data exchange under the DAVI framework. https://www.transportation.gov/av/data/wzdx The FHWA CARMA Program is developing and encouraging collaboration around an open source CARMASM Platform for investigation of interactions between AVs and transportation infrastructure. Documentation and code are available in its GitHub repository. https://cms7.fhwa.dot.gov/research/research-programs/operations/carma-overview

State Regulatory Perspectives

Many States have passed legislation or issued executive orders regarding the deployment, testing, or operations of autonomous vehicles within their jurisdictions. The National Conference of State Legislatures (NCSL) maintains a database of this legislation. https://www.ncsl.org/research/transportation/autonomous-vehicles-self-driving-vehicles-enacted-legislation.aspx

Applicable Standards

As indicated above, the USDOT AV 3.0 document provides an extensive list of standards and standards development efforts relevant to AVs and CAT. Published standards may be obtained directly from the particular standards development organization (SDO).

H IEEE: https://www.ieee.org/standards/index.html

H SAE International: https://www.sae.org/standards/

H ISO: https://www.iso.org/standards.html

H AASHTO: http://www.aashtoresource.org/standards

H NTCIP: https://www.ntcip.org/document-numbers-and-status/