IEEE INTELLIGENT VEHICLES SYMPOSIUM
Anchorage, Alaska, USA June 4 – 7, 2023
SceGen: Workshop on Scenario Generation for Testing
Autonomous Vehicles (WS10)
About
Autonomous vehicles (AVs) must be rigorously tested for safety before being deployed in the real world. Scenario-based testing of AVs is a promising approach since it evaluates the response of the complete AV system. Significant research has been conducted in this area. Scenarios are an essential form of data to represent environments in simulation and the real world. Simulation testing provides a faster feedback loop to AV developers since it is highly accessible, controllable, and cheaper to run tests in relative to the real world. In a simulation, to provide a comprehensive evaluation of the safety of an AV, the AV would need to be tested on a large number of desirable scenarios. Desirable scenarios would typically expose hazardous or risky AV responses in the simulation itself. Scenarios broadly contain static components (such as roads and roadside objects) and dynamic components (such as vehicle and pedestrian behaviors). Further, they are broadly defined at varying levels of abstraction. There’s a need to represent them in a 3D photorealistic environment to evaluate perception responses and in physics-based environments to evaluate vehicle control responses. This has been addressed in the development of several game engine-based simulators such as CARLA and BeammNG.tech. Scenario standards are currently being developed (such as in the case of OpenDRIVE and OpenSCENARIO) and iterated upon.
While the field has primarily settled on terminology and initial standards existing for roads and scenarios, the current emphasis is shifting to creating scenarios from naturalistic datasets, adding greater dynamism to scenarios via interactive NPC cars and pedestrian models, and using procedural content generation methods. Further, much work remains to represent the astonishing diversity of roads, pedestrian and driver behavior, and situations encountered across the globe. Significant discussion is required in the design of the aforementioned models and processes.
Topics
Scenario modeling, extraction and reconstruction from naturalistic data.
Data standards and datasets for roads and scenarios.
Taxonomies and Ontologies for scenario generation.
Dynamic scenario generation - vehicles and pedestrian behaviors, weather and other object dynamics.
Procedural content generation algorithms.
Augmenting datasets with synthetic scenario data.
Machine Learning for scenario generation.
Adversarial scenario generation.
Tools for scenario generation.
Programming languages for scenario generation.
AV stacks as systems under test and candidates for scenario generation methods.
Metrics for evaluating scenarios.
Schedule and Location
We start the workshop with invited talks and end with a panel discussion covering research needs, open questions, current progress, and future scope of work. Each invited talk will be 30 minutes, including a 20-minute presentation and a 10-minute QA.
The workshop will be hybrid. Zoom links will be sent out to attendees. Each accepted paper will need at least one of the authors present in the workshop.
Below is the schedule in Alaska Time, USA (GMT - 8h). The workshop starts at 13:40 p.m. and ends at 17:50 p.m. The floorplan will be shared soon.
Submission Guidelines
Paper submission link: https://edas.info/newPaper.php?c=30459&track=115618
The workshop code is WS10 and the title is "WS10 - Workshop on Scenario Generation for Testing Autonomous Vehicles" in the submission page.
Important Deadlines:
February 01, 2023: Workshop Paper Submission Deadline (firm deadline, no extension)
March 30, 2023: Workshop Paper Notification of Acceptance
April 22, 2023: Workshop Final Paper Submission Deadline
Paper Format:
Submitted papers shall not exceed six pages (two additional pages allowed with a fee) as a PDF file in IEEE two-column format. All presented papers will be published by the IEEE, and the conference proceedings will be submitted to the IEEE Xplore digital library, as long they follow the same review process of IEEE IV 2023 so that each paper will undergo a peer-reviewing process by at least two members of the International Program Committee. Contributions will be reviewed according to relevance, originality and novel ideas, technical soundness, and quality of presentation. The templates for papers can be found here: https://www.ieee.org/conferences/publishing/templates.html
Accepted Contributions:
Regular papers: with max six pages (two additional pages allowed with a fee)
Extended abstracts with new ideas: The new ideas are about innovative, forward-looking and even provoking ideas described in (short) papers that present solid arguments about future directions or new techniques that might not have been yet validated thoroughly (max 2/4 pages)
Tool papers: describing existing software packages and tools at various maturity levels that seek to bridge the gap between research and practice (max 4 pages).
Case studies and empirical research papers are also welcome. (max 4 pages)
Registration
Free workshops-day only registrations for the invited speakers.
Free workshops-day only registrations for IEEE ITSS members. You can register for a small membership fee compared to the workshop fee. Here goes the link https://ieee-itss.org/
Paid attendance for everyone else. You will only need to pay for the workshop day only if you are attending the workshop.
Invited Speakers
Prof Gustav Markkula
University of Leeds, UK
Prof. Gustav Markkula is an engineer by training, and applies quantitative methods and models to the study of human behaviour and cognition in road traffic. He has a background in automotive industry R&D (Volvo), and is currently Chair in Applied Behaviour Modelling at the Institute for Transport Studies, University of Leeds, UK. In his research, he specialises in the adoption and integration of models from computational cognitive neuroscience, to support development and testing of safe and human-acceptable technology and automation.
Prof Leilani Gilpin
University of California, Santa Cruz
Leilani H. Gilpin is an Assistant Professor in the Department of Computer Science and Engineering at UC Santa Cruz. Her research focuses on the design and analysis of methods for autonomous systems to explain themselves. Her work has applications to robust decision-making, system debugging, and accountability. She holds a PhD in Computer Science from MIT, an M.S. in Computational and Mathematical Engineering from Stanford University, and a B.S. in Mathematics (with honors), B.S. in Computer Science (with highest honors), and a music minor from UC San Diego. Outside of research, Leilani enjoys swimming, cooking, hiking, and org-mode.
Dr. Chrysanthi Papamichail
Lead Research Software Engineer, BeamNG
Dr Chrysanthi Papamichail contributes currently to the simulation industry (BeamNG GmbH) as Lead Research and Software engineer. Her activity aims to develop further and validate the company's simulator to meet the needs of relevant academic research and industrial projects. She has industrial experience in R&D on sensor data fusion and perception for ADAS/AD technologies, for car manufacturer (Renault–Nissan–Mitsubishi Alliance). Her doctoral research was in Applied Mathematics & Statistics, for stochastic models of mechanical engineering problems (Université de Technologie de Compiègne).
Lukas Birkemeyer
Technische Universität Braunschweig
Lukas Birkemeyer is a PhD student and researcher at the Technische Universität Braunschweig in Germany. He is part of the “Responsible AI in the Digital Society” program, funded by the Ministry for Science and Culture of Lower Saxony. Lukas has a background in electrical engineering and information technology and has worked in the automotive industry. His research focuses on verifying and validating ADAS and ADS. He aims to develop a systematic approach for generating scenarios in line with the SOTIF standard (ISO 21448) to ensure the reliability and safety of these systems. Lukas’ work is crucial in advancing technology in the automotive industry.
Organizing Committee
Prof. Jim Whitehead
University of California, Santa Cruz
Prof. Alessio Gambi
IMC University of Applied Science Krems
Golam Md Muktadir
University of California, Santa Cruz
Ishaan Paranjape
University of California, Santa Cruz
Abdul Jawad
University of California, Santa Cruz
For more information or any questions please contact Ishaan Paranjape (iparanja@ucsc.edu).