DOCSLIB.ORG

INTELIGENCIA EN REDES DE COMUNICACIONES Trabajos De La Asignatura Curso 11-12

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
INTELIGENCIA EN REDES DE COMUNICACIONES Trabajos De La Asignatura Curso 11-12 INTELIGENCIA EN REDES DE COMUNICACIONES Trabajos de la asignatura Curso 11-12 Departamento de Ingeniería Telemática Universidad Carlos III de Madrid INTELIGENCIA EN REDES DE COMUNICACIONES Trabajos de la asignatura Curso 11-12 Julio Villena Román Raquel M. Crespo García Departamento de Ingeniería Telemática Universidad Carlos III de Madrid {jvillena, rcrespo}@it.uc3m.es ÍNDICE 1. El coche inteligente. .................................................................. 1 Daniel Asegurado Turón Si alguna vez has soñado con tener un coche como KITT, el coche fantástico, ese sueño podría hacerse realidad en un futuro no muy lejano. El sueño del coche fantástico es una realidad que está emergiendo con fuerza en varios países, incluido España. El objetivo del presente documento es realizar un recorrido por la historia del automóvil hasta llegar a los considerados coches inteligentes, donde nos detendremos a analizar los diferentes proyectos que trabajan con esta idea. Palabras clave: Coche autónomo, automatización de vehículos, autoconducción, Mecatrónica, radar, LIDAR, GPS, visión artificial, eye tracking, estereoscopía. 2. Sistemas expertos: MYCIN ...................................................... 11 Ainhoa Sesmero Fernández, Sandra Pinero Sánchez En este documento se explica lo que es un sistema experto, sus principales usos y en concreto hablaremos del sistema MYCIN. Uno de los primeros sistemas expertos para la detección y tratamiento de enfermedades de la sangre. Palabras clave: Sistemas expertos (SE), aplicaciones en medicina, MYCIN, medicina, organismos, conocimiento. - i - 3. Diseño e implementación de un agente inteligente Mario A.I. .. 19 Yuchen Du, Virginia Izquierdo Bermúdez En este artículo se detalla el diseño y la implementación de un agente inteligente capaz de jugar al videojuego de Mario así cómo las técnicas utilizadas para este diseño. Palabras clave: Árbol de decisión, agente inteligente, Mario, enemigos, agente, inteligencia, escenario. 4. Jugador virtual de póker Texas Hold’em ................................. 29 Alberto Álvarez Polegre, Lucas Bernalte Pérez, David Díaz Chinea El póquer es un juego de cartas de los llamados de "apuestas", en los que los jugadores, con todas o parte de sus cartas ocultas, hacen apuestas sobre una puja inicial, recayendo la suma total de las apuestas en el jugador o jugadores con la mejor combinación de cartas. Este trabajo hace un diseño de un jugador virtual de póker Texas Hold’em. Palabras clave: Póker, Texas Hold’em, probabilidad, jugada, mano, estrategia. 5. Inteligencia artificial en el mundo de la aviación ..................... 37 Daniel Ruesga Ibañez, Hector Molla Ruiz En este documento vamos a realizar un estudio del impacto que ha tenido la inteligencia artificial en el mundo de la aviación moderna. Nos centraremos de forma especial en los sistemas de pilotaje automático disponibles en los aviones y su avance hasta los vehículos no tripulados (UAVS). Palabras clave: Sistemas expertos, aviación, piloto automático, navegación, vuelo, UAV. 6. Sistemas Expertos en meteorología ......................................... 45 Carlos Lentisco Sánchez, Tania Romero Rodríguez Las aplicaciones basadas en inteligencia artificial son innumerables y van desde la biomedicina o educación hasta la predicción de las condiciones meteorológicas que realizan los sistemas expertos, objeto de estudio y razón de ser de este documento. Esta es la razón que justifica la búsqueda del ser humano por predecir fiablemente las condiciones meteorológicas. A lo largo de este documento nos introduciremos más en los sistemas expertos repasando su evolución histórica hasta nuestros días para abordar posteriormente y con mayor profundidad un ejemplo de estos sistemas. Centrándonos en el sistema Prometeo. Palabras clave: Inteligencia, meteorología, sistemas expertos, predicción meteorológica. - ii - 7. Traducción automática. Técnicas y aplicaciones ...................... 53 Isabel Cuadrado Gutiérrez, David Ferrer Figueroa En este texto se aborda un tema de gran relevancia en la actualidad, el concepto de la traducción automática. Se analizará esta área desde su origen hasta la actualidad, tratando la evolución de las distintas técnicas empleadas para obtener la mayor calidad posible en las traducciones. Asimismo, en el trrabajo se hace un breve repaso a las aplicaciones más utilizadas y al desarrollo futuro previsto en este campo teniendo en cuenta, entre otros factores, la rentabilidad del negocio y la calidad de la traducción. Palabras clave: Inteligencia artificial, Traducción automática, algoritmo de corpus. 8. Resolución de Puzzles Deslizantes ........................................... 61 Lara Rojo Celemín, Iván Suárez Morejudo En 1878, Sam Loyd daba un premio de $1.000 a quien fuera capaz de resolver su famoso 14-15 Puzzle. Se trataba de un puzzle deslizante de 16 piezas que no tenía solución, lo que hizo que muchas personas se volvieran locas tratando de resolverlo. En este trabajo tratamos de presentar diferentes algoritmos para resolver el 15-Puzzle desde el punto de vista de la Inteligencia Artificial. Para ello, utilizamos el algoritmo de búsqueda en árboles A*, del que daremos una visión general que luego particularizaremos a nuestro caso del 15-Puzzle. Aunque en este trabajo sólo nos hemos centrado en resolver el rompecabezas de 16 piezas, los algoritmos explicados pueden utilizarse para resolver cualquier puzzle deslizante n x n con solución. Palabras clave: Puzzles, Algoritmos, permutaciones pares, heurística, Inteligencia Artificial, resolución humana, puzzles deslizantes, 15-puzzle, Sam Loyd, Algoritmo A*, herística, Turing. 9. Desarrollo de un agente autónomo jugador de Super Mario .... 70 Adrián Amor Martín, Cristina García Muñoz En este documento se va a detallar cómo realizar un agente, basado en árboles de decisión, que sea capaz jugar de forma autónoma al juego Super Mario en el marco del concurso Mario AI 2011. Palabras clave: Agente, árboles de decisión, inteligencia artificial, Super Mario. 10. Implementación de un Agente para un jugador de Mario A.I. .. 78 Alejandro Pérez Cruz, David de la Horra Iglesias En este documento presentamos una posible implementación de un agente que pueda jugar al juego SuperMario Bross. con una cierta inteligencia autónoma. - iii - Palabras clave: Algoritmos de búsqueda, lenguaje de programación Java, diseño e implementación de un agente, búsqueda, inteligencia artificial, agentes. 11. Robot de Robocode ................................................................. 86 Felipe Llinares López, Juan José Torres García En esta memoria describimos una implementación que hemos realizado de un robot de competición para el programa Robocode. Palabras clave: Algoritmos, Rendimiento, Diseño, Robocode. 12. Minería de Datos .................................................................... 94 Óscar Palomo Miñambres En este artículo analizaremos las ideas básicas que sustentan el Data Mining y, más concretamente, la utilización de árboles de decisión como herramienta estadística avanzada. Presento también un ejemplo ficticio de la aplicación de estas técnicas: predecir clientes que tienen un alto porcentaje de impagos. Palabras clave: Minería de Datos, árboles de decisión, impagos, predicción. - iv - El coche inteligente En busca de mayor seguridad, sostenibilidad y confort Daniel Asegurado Turón Universidad Carlos III de Madrid Leganés, Madrid, España [email protected] RESUMEN Durante muchos años, esta idea del automóvil no ha sido más que Si alguna vez has soñado con tener un coche como KITT, el coche mera ciencia ficción. Sin embargo, la realidad hoy es bien fantástico, ese sueño podría hacerse realidad en un futuro no muy diferente. En los casi treinta años que separan la emisión de la lejano. El sueño del coche fantástico es una realidad que está legendaria serie de la actualidad, se han realizado grandes emergiendo con fuerza en varios países, incluido España. El progresos en el estudio del coche inteligente. Organismos objetivo del presente documento es realizar un recorrido por la oficiales, como la Unión Europea, fabricantes de automóviles, historia del automóvil hasta llegar a los considerados coches como Volvo, o incluso compañías de otros sectores ajenos al inteligentes, donde nos detendremos a analizar los diferentes motor, como es el caso de Google, se han lanzado a la carrera por proyectos que trabajan con esta idea. conseguir el coche del futuro. ¿Cuál es el interés por obtener un coche de estas características? Palabras clave Siguiendo con la analogía del coche fantástico, el principal interés Coche autónomo, automatización de vehículos, autoconducción, sigue siendo salvar vidas. Mientras que KITT se preocupaba de Mecatrónica, radar, LIDAR, GPS, visión artificial, eye tracking, salvarle la vida a su compañero Michael, el objetivo de los estereoscopía organismos institucionales como la Unión Europea es reducir el número de muertes al volante. Según la DGT [1], en el primer semestre de 2011 ya se contabilizaban 986 víctimas mortales por 1. INTRODUCCION accidentes en la carretera (no están incluidas las víctimas de Todos hemos tenido envidia del mítico Michael Knight cuando, lesiones graves o leves en el total de accidentes). En cifras con sólo una orden de voz llamaba a KITT, su coche fantástico, relativas, en Europa murieron 62 personas por millón de para que le fuese a recoger allá donde se encontrase. KITT no sólo habitantes el pasado 2010, según el informe del ETSC sobre era capaz de conducir autónomamente, sino que era
Load more

Recommended publications
  • Rapport AI Internationale Verkenning Van De Sociaaleconomische
    RAPPORT ARTIFICIËLE INTELLIGENTIE INTERNATIONALE VERKENNING VAN DE SOCIAAL-ECONOMISCHE IMPACT STAND VAN ZAKEN EIND OKTOBER 2020 5. AI als een octopus met tentakels in diverse sectoren: een illustratie Artificiële intelligentie: sociaal-economische verkenning 5 AI als een octopus met tentakels in diverse sectoren: een illustratie 5.1 General purpose: Duurzame ontwikkeling (SDG’s) AI is een technologie voor algemene doeleinden en kan worden aangewend om het algemeen maatschappelijk welzijn te bevorderen.28 De bijdrage van AI in het realiseren van de VN- doelstellingen voor duurzame ontwikkeling (SDG's) op het vlak van onder meer onderwijs, gezondheidszorg, vervoer, landbouw en duurzame steden, vormt hiervan het beste bewijs. Veel openbare en particuliere organisaties, waaronder de Wereldbank, een aantal agentschappen van de Verenigde Naties en de OESO, willen AI benutten om stappen vooruit te zetten op de weg naar de SDG's. Uit de database van McKinsey Global Institute29 met AI-toepassingen blijkt dat AI ingezet kan worden voor elk van de 17 duurzame ontwikkelingsdoelstellingen. Onderstaande figuur geeft het aantal AI-use cases (toepassingen) in de database van MGI aan die elk van de SDG's van de VN kunnen ondersteunen. Figuur 1: AI use cases voor de ondersteuning van de SDG’s30 28 Steels, L., Berendt, B., Pizurica, A., Van Dyck, D., Vandewalle,J. Artificiële intelligentie. Naar een vierde industriële revolutie?, Standpunt nr. 53 KVAB, 2017. 29 McKinsey Global Institute, notes from the AI frontier applying AI for social good, Discussion paper, December 2018. Chui, M., Chung, R., van Heteren, A., Using AI to help achieve sustainable development goals, UNDP, January 21, 2019 30 Noot McKinsey: This chart reflects the number and distribution of use cases and should not be read as a comprehensive evaluation of AI potential for each SDG; if an SDG has a low number of cases, that is a reflection of our library rather than of AI applicability to that SDG.
    [Show full text]
  • Appearance-Based Gaze Estimation with Deep Learning: a Review and Benchmark
    MANUSCRIPT, APRIL 2021 1 Appearance-based Gaze Estimation With Deep Learning: A Review and Benchmark Yihua Cheng1, Haofei Wang2, Yiwei Bao1, Feng Lu1,2,* 1State Key Laboratory of Virtual Reality Technology and Systems, SCSE, Beihang University, China. 2Peng Cheng Laboratory, Shenzhen, China. fyihua c, baoyiwei, [email protected], [email protected] Abstract—Gaze estimation reveals where a person is looking. It is an important clue for understanding human intention. The Deep learning algorithms recent development of deep learning has revolutionized many computer vision tasks, the appearance-based gaze estimation is no exception. However, it lacks a guideline for designing deep learn- ing algorithms for gaze estimation tasks. In this paper, we present a comprehensive review of the appearance-based gaze estimation methods with deep learning. We summarize the processing Chip pipeline and discuss these methods from four perspectives: deep feature extraction, deep neural network architecture design, personal calibration as well as device and platform. Since the Platform Calibration Feature Model data pre-processing and post-processing methods are crucial for gaze estimation, we also survey face/eye detection method, data rectification method, 2D/3D gaze conversion method, and gaze origin conversion method. To fairly compare the performance of various gaze estimation approaches, we characterize all the Fig. 1. Deep learning based gaze estimation relies on simple devices and publicly available gaze estimation datasets and collect the code of complex deep learning algorithms to estimate human gaze. It usually uses typical gaze estimation algorithms. We implement these codes and off-the-shelf cameras to capture facial appearance, and employs deep learning set up a benchmark of converting the results of different methods algorithms to regress gaze from the appearance.
    [Show full text]
  • Through the Eyes of the Beholder: Simulated Eye-Movement Experience (“SEE”) Modulates Valence Bias in Response to Emotional Ambiguity
    Emotion © 2018 American Psychological Association 2018, Vol. 18, No. 8, 1122–1127 1528-3542/18/$12.00 http://dx.doi.org/10.1037/emo0000421 BRIEF REPORT Through the Eyes of the Beholder: Simulated Eye-Movement Experience (“SEE”) Modulates Valence Bias in Response to Emotional Ambiguity Maital Neta and Michael D. Dodd University of Nebraska—Lincoln Although some facial expressions provide clear information about people’s emotions and intentions (happy, angry), others (surprise) are ambiguous because they can signal both positive (e.g., surprise party) and negative outcomes (e.g., witnessing an accident). Without a clarifying context, surprise is interpreted as positive by some and negative by others, and this valence bias is stable across time. When compared to fearful expressions, which are consistently rated as negative, surprise and fear share similar morphological features (e.g., widened eyes) primarily in the upper part of the face. Recently, we demonstrated that the valence bias was associated with a specific pattern of eye movements (positive bias associated with faster fixation to the lower part of the face). In this follow-up, we identified two participants from our previous study who had the most positive and most negative valence bias. We used their eye movements to create a moving window such that new participants viewed faces through the eyes of one our previous participants (subjects saw only the areas of the face that were directly fixated by the original participants in the exact order they were fixated; i.e., Simulated Eye-movement Experience). The input provided by these windows modulated the valence ratings of surprise, but not fear faces.
    [Show full text]
  • Modulation of Saccadic Curvature by Spatial Memory and Associative Learning
    Modulation of Saccadic Curvature by Spatial Memory and Associative Learning Dissertation zur Erlangung des Doktorgrades der Naturwissenschaften (Dr. rer. nat.) dem Fachbereich Psychologie der Philipps-Universität Marburg vorgelegt von Stephan König aus Arnsberg Marburg/Lahn, Oktober 2010 Vom Fachbereich Psychologie der Philipps-Universität Marburg als Dissertation angenommen am _________________ Erstgutachter: Prof. Dr. Harald Lachnit, Philipps Universität Marburg Zweitgutachter: Prof. Dr. John Pearce, Cardiff University Tag der mündlichen Prüfung am: 27.10.2010 Acknowledgment I would like to thank Prof. Dr. Harald Lachnit for providing the opportunity for my work on this thesis. I would like to thank Prof. Dr. Harald Lachnit, Dr. Anja Lotz and Prof. Dr. John Pearce for proofreading, reviewing and providing many valuable suggestions during the process of writing. I would like to thank Peter Nauroth, Sarah Keller, Gloria-Mona Knospe, Sara Lucke and Anja Riesner who helped in recruiting participants as well as collecting data. I would like to thank the Deutsche Forschungsgemeinschaft for their support through the graduate program NeuroAct (DFG 885/1), and project Blickbewegungen als Indikatoren assoziativen Lernens (DFG LA 564/20-1) granted to Prof. Dr. Harald Lachnit. Summary The way the eye travels during a saccade typically does not follow a straight line but rather shows some curvature instead. Converging empirical evidence has demonstrated that curvature results from conflicting saccade goals when multiple stimuli in the visual periphery compete for selection as the saccade target (Van der Stigchel, Meeter, & Theeuwes, 2006). Curvature away from a competing stimulus has been proposed to result from the inhibitory deselection of the motor program representing the saccade towards that stimulus (Sheliga, Riggio, & Rizzolatti, 1994; Tipper, Howard, & Houghton, 2000).
    [Show full text]
  • A Survey of Autonomous Driving: Common Practices and Emerging Technologies
    Accepted March 22, 2020 Digital Object Identifier 10.1109/ACCESS.2020.2983149 A Survey of Autonomous Driving: Common Practices and Emerging Technologies EKIM YURTSEVER1, (Member, IEEE), JACOB LAMBERT 1, ALEXANDER CARBALLO 1, (Member, IEEE), AND KAZUYA TAKEDA 1, 2, (Senior Member, IEEE) 1Nagoya University, Furo-cho, Nagoya, 464-8603, Japan 2Tier4 Inc. Nagoya, Japan Corresponding author: Ekim Yurtsever (e-mail: [email protected]). ABSTRACT Automated driving systems (ADSs) promise a safe, comfortable and efficient driving experience. However, fatalities involving vehicles equipped with ADSs are on the rise. The full potential of ADSs cannot be realized unless the robustness of state-of-the-art is improved further. This paper discusses unsolved problems and surveys the technical aspect of automated driving. Studies regarding present challenges, high- level system architectures, emerging methodologies and core functions including localization, mapping, perception, planning, and human machine interfaces, were thoroughly reviewed. Furthermore, many state- of-the-art algorithms were implemented and compared on our own platform in a real-world driving setting. The paper concludes with an overview of available datasets and tools for ADS development. INDEX TERMS Autonomous Vehicles, Control, Robotics, Automation, Intelligent Vehicles, Intelligent Transportation Systems I. INTRODUCTION necessary here. CCORDING to a recent technical report by the Eureka Project PROMETHEUS [11] was carried out in A National Highway Traffic Safety Administration Europe between 1987-1995, and it was one of the earliest (NHTSA), 94% of road accidents are caused by human major automated driving studies. The project led to the errors [1]. Against this backdrop, Automated Driving Sys- development of VITA II by Daimler-Benz, which succeeded tems (ADSs) are being developed with the promise of in automatically driving on highways [12].
    [Show full text]
  • Passenger Response to Driving Style in an Autonomous Vehicle
    Passenger Response to Driving Style in an Autonomous Vehicle by Nicole Belinda Dillen A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Master of Mathematics in Computer Science Waterloo, Ontario, Canada, 2019 c Nicole Dillen 2019 I hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners. I understand that my thesis may be made electronically available to the public. ii Abstract Despite rapid advancements in automated driving systems (ADS), current HMI research tends to focus more on the safety driver in lower level vehicles. That said, the future of automated driving lies in higher level systems that do not always require a safety driver to be present. However, passengers might not fully trust the capability of the ADS in the absence of a safety driver. Furthermore, while an ADS might have a specific set of parameters for its driving profile, passengers might have different driving preferences, some more defensive than others. Taking these preferences into consideration is, therefore, an important issue which can only be accomplished by understanding what makes a passenger uncomfortable or anxious. In order to tackle this issue, we ran a human study in a real-world autonomous vehicle. Various driving profile parameters were manipulated and tested in a scenario consisting of four different events. Physiological measurements were also collected along with self- report scores, and the combined data was analyzed using Linear Mixed-Effects Models. The magnitude of a response was found to be situation dependent: the presence and proximity of a lead vehicle significantly moderated the effect of other parameters.
    [Show full text]
  • Gaze Shifts As Dynamical Random Sampling
    GAZE SHIFTS AS DYNAMICAL RANDOM SAMPLING Giuseppe Boccignone Mario Ferraro Dipartimento di Scienze dell’Informazione Dipartimento di Fisica Sperimentale Universita´ di Milano Universita´ di Torino Via Comelico 39/41 via Pietro Giuria 1 20135 Milano, Italy 10125 Torino, Italy [email protected] [email protected] ABSTRACT so called “focus of attention” (FOA), to extract detailed in- formation from the visual environment. An average of three We discuss how gaze behavior of an observer can be simu- eye fixations per second generally occurs, intercalated by lated as a Monte Carlo sampling of a distribution obtained saccades, during which vision is suppressed. The succession from the saliency map of the observed image. To such end we of gaze shifts is referred to as a scanpath [1]. propose the Levy Hybrid Monte Carlo algorithm, a dynamic Beyond the field of computational vision,[2], [3],[4], [5] Monte Carlo method in which the walk on the distribution and robotics [6], [7], much research has been performed in landscape is modelled through Levy flights. Some prelimi- the image/video coding [8], [9]. [10], [11] , [12], and im- nary results are presented comparing with data gathered by age/video retrieval domains [13],[14]. eye-tracking human observers involved in an emotion recog- However, one issue which is not addressed by most mod- nition task from facial expression displays. els [15] is the ”noisy”, idiosyncratic variation of the random Index Terms— Gaze analysis, Active vision, Hybrid exploration exhibited by different observers when viewing the Monte Carlo, Levy flights same scene, or even by the same subject along different trials [16] (see Fig.
    [Show full text]
  • New Methodology to Explore the Role of Visualisation in Aircraft Design Tasks
    220 Int. J. Agile Systems and Management, Vol. 7, Nos. 3/4, 2014 New methodology to explore the role of visualisation in aircraft design tasks Evelina Dineva*, Arne Bachmann, Erwin Moerland, Björn Nagel and Volker Gollnick Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR), German Aerospace Center, Blohmstr. 18, 21079 Hamburg, Germany E-mail: [email protected] E-mail: [email protected] E-mail: [email protected] E-mail: [email protected] E-mail: [email protected] *Corresponding author Abstract: A toolbox for the assessment of engineering performance in a realistic aircraft design task is presented. Participants solve a multidisciplinary optimisation (MDO) task by means of a graphical user interface (GUI). The quantity and quality of visualisation may be systematically varied across experimental conditions. The GUI also allows tracking behavioural responses like cursor trajectories and clicks. Behavioural data together with evaluation of the generated aircraft design can help uncover details about the underlying decision making process. The design and the evaluation of the experimental toolbox are presented. Pilot studies with two novice and two advanced participants confirm and help improve the GUI functionality. The selection of the aircraft design task is based on a numerical analysis that helped to identify a ‘sweet spot’ where the task is not too easy nor too difficult. Keywords: empirical performance evaluation; visualisation; aircraft design; multidisciplinary optimisation; MDO; multidisciplinary collaboration; collaborative engineering. Reference to this paper should be made as follows: Dineva, E., Bachmann, A., Moerland, E., Nagel, B. and Gollnick, V. (2014) ‘New methodology to explore the role of visualisation in aircraft design tasks’, Int.
    [Show full text]
  • Human Visual Attention Prediction Boosts Learning & Performance of Autonomous Driving
    Human Visual Attention Prediction Boosts Learning & Performance of Autonomous Driving Agents Alexander Makrigiorgos, Ali Shafti, Alex Harston, Julien Gerard, and A. Aldo Faisal Abstract— Autonomous driving is a multi-task problem re- quiring a deep understanding of the visual environment. End- to-end autonomous systems have attracted increasing interest as a method of learning to drive without exhaustively pro- gramming behaviours for different driving scenarios. When humans drive, they rely on a finely tuned sensory system which enables them to quickly acquire the information they need while filtering unnecessary details. This ability to identify task- specific high-interest regions within an image could be beneficial to autonomous driving agents and machine learning systems in general. To create a system capable of imitating human gaze patterns and visual attention, we collect eye movement data from human drivers in a virtual reality environment. We use this data to train deep neural networks predicting Fig. 1: A subject driving in the VR simulation. where humans are most likely to look when driving. We then use the outputs of this trained network to selectively mask intuitive human-robot interfaces [18]–[21]. The ability to driving images using a variety of masking techniques. Finally, understand and recreate this attention-focusing mechanism autonomous driving agents are trained using these masked has the potential to provide similar benefits to machine images as input. Upon comparison, we found that a dual-branch learning systems which have to parse complex visual scenes architecture which processes both raw and attention-masked images substantially outperforms all other models, reducing in order to complete their tasks, reducing computational bur- error in control signal predictions by 25.5% compared to a den and improving performance by quickly identifying key standard end-to-end model trained only on raw images.
    [Show full text]
  • DOT's Efforts to Streamline Technology Transfer
    Deputy Assistant Secretary 1200 New Jersey Avenue, S.E. U.S. Department for Research and Technology Washington, DC 20590 of Transportation Office of the Secretary of Transportation May 14, 2019 TO: Dr. Michael Walsh, Technology Partnerships Office National Institute of Standards and Technology FROM: Diana Furchtgott-Roth Deputy Assistant Secretary for Research and Technology SUBJECT: Fiscal Year 2018 Technology Transfer (T2) Annual Performance Report Every year, the Department of Commerce (DOC) submits a Federal Laboratory T2 Fiscal Year Summary Report to the President and the Congress in accordance with 15 U.S.C. 3710(g)(2). The report summarizes the implementation of technology transfer authorities established by the Technology Transfer Commercialization Act of 2000 (Pub. L. 106-404) and other legislation. This report summarizes U.S. DOT’s information for DOC’s Fiscal Year 2018 Summary Report. Please submit questions pertaining to this report to Santiago Navarro at [email protected] or 202-366-0849. Attachment: Fiscal Year 2018 Technology Transfer (T2) Annual Performance Report. This page is intentionally left blank. Technology Transfer Report FY 2018 This page is intentionally left blank. U.S. Department of Transportation Office of the Assistant Secretary for Research and Technology MAY 2019 Table of Contents List of Figures ...................................................................................................................................ii List of Tables ....................................................................................................................................ii
    [Show full text]
  • THROUGH the HANDOFF LENS: COMPETING VISIONS of AUTONOMOUS FUTURES Jake Goldenfein†, Deirdre K
    THROUGH THE HANDOFF LENS: COMPETING VISIONS OF AUTONOMOUS FUTURES Jake Goldenfein†, Deirdre K. Mulligan††, Helen Nissenbaum††† & Wendy Ju‡ ABSTRACT The development of autonomous vehicles is often presented as a linear trajectory from total human control to total autonomous control, with only technical and regulatory hurdles in the way. But below the smooth surface of innovation-speak lies a battle over competing autonomous vehicle futures with ramifications well beyond driving. Car companies, technology companies, and others are pursuing alternative autonomous vehicle visions, and each involves an entire reorganization of society, politics, and values. Instead of subscribing to the story of inevitable linear development, this paper explores three archetypes of autonomous vehicles—advanced driver-assist systems, fully driverless cars, and connected cars—and the futures they foretell as the ideal endpoints for different classes of actors. We introduce and use the Handoff Model—a conceptual model for analyzing the political and ethical contours of performing a function with different configurations of human and technical actors—in order to expose the political and social reconfigurations intrinsic to those different futures. Using the Handoff Model, we analyze how each archetype both redistributes the task of “driving” across different human and technical actors and imposes political and ethical propositions both on human “users” and society at large. The Handoff Model exposes the baggage each transport model carries and serves as a guide in identifying the desirable and necessary technical, legal, and social dynamics of whichever future we choose. DOI: https://doi.org/10.15779/Z38CR5ND0J © 2020 Jake Goldenfein, Deirdre K. Mulligan, Helen Nissenbaum & Wendy Ju.
    [Show full text]
  • Open-Source Software for Real-Time Pervasive Head-Mounted Eye Tracking
    EyeRecToo: Open-source Software for Real-time Pervasive Head-mounted Eye Tracking Thiago Santini, Wolfgang Fuhl, David Geisler and Enkelejda Kasneci Perception Engineering, University of Tubingen,¨ Tubingen,¨ Germany thiago.santini, wolfgang.fuhl, david.geisler, enkelejda.kasneci @uni-tuebingen.de { } Keywords: Eye Movements, Pupil Detection, Calibration, Gaze Estimation, Open-source, Eye Tracking, Data Acquisi- tion, Human-computer Interaction, Real-time, Pervasive. Abstract: Head-mounted eye tracking offers remarkable opportunities for research and applications regarding pervasive health monitoring, mental state inference, and human computer interaction in dynamic scenarios. Although a plethora of software for the acquisition of eye-tracking data exists, they often exhibit critical issues when pervasive eye tracking is considered, e.g., closed source, costly eye tracker hardware dependencies, and requir- ing a human supervisor for calibration. In this paper, we introduce EyeRecToo, an open-source software for real-time pervasive head-mounted eye-tracking. Out of the box, EyeRecToo offers multiple real-time state-of- the-art pupil detection and gaze estimation methods, which can be easily replaced by user implemented algo- rithms if desired. A novel calibration method that allows users to calibrate the system without the assistance of a human supervisor is also integrated. Moreover, this software supports multiple head-mounted eye-tracking hardware, records eye and scene videos, and stores pupil and gaze information, which are also available as a real-time stream. Thus, EyeRecToo serves as a framework to quickly enable pervasive eye-tracking research and applications. Available at: www.ti.uni-tuebingen.de/perception. 1 INTRODUCTION and application developers have practically no direct alternatives if the system does not work under the re- In the past two decades, the number of researchers quired conditions (Santini et al., 2016b).
    [Show full text]