Event Actions
Abstract:
Safety is foundational to the successful deployment of autonomous vehicles. By integrating model-based and data-driven approaches, automated agents can leverage both structured knowledge and real-world data, leading to more robust decision-making. Building a hierarchical architecture with system 1 and system 2 layers, and embedding safety at each layer, enable the autonomous vehicles to remain aware of their physical constraints within the environment. This ensures that strategic decisions are made with caution and consideration for long-term safety, while immediate reactive actions are prioritized for real-time hazard avoidance. It also facilitates more effective testing and validation.
In this talk, I will share my insights and approaches for embedding safety at each layer of autonomous vehicle design, drawing on both model-based and data-driven methods.
Bio:
H. Eric Tseng received the B.S. degree from the National Taiwan University in 1986, and the M.S. and Ph.D. degrees in mechanical engineering from the University of California, Berkeley in 1991 and 1994, respectively. In May 2024, he joined the University of Texas at Arlington as a distinguished university professor in the Department of Electrical Engineering.
In 1994, he joined Ford Motor Company. At Ford (1994-2022), he had a productive career and retired as a Senior Technical Leader of Controls and Automated Systems in Research and Advanced Engineering. Many of his contributed technologies led to production vehicles implementation. His technical achievements have been honored with Ford's annual technology award, the Henry Ford Technology Award, on seven occasions. Additionally, he was the recipient of the Control Engineering Practice Award from the American Automatic Control Council in 2013, and the recipient of the Soichiro Honda Medal from the American Society of Mechanical Engineering in 2024. He has over 100 U.S. patents and over 160 publications. He is a member of the National Academy of Engineering as of 2021.