Competent Driving Behavior
Date and Time: Tuesday, July 11, 2023: 1:30 PM - 5:00 PM
Presentation Description
Automated vehicles (AVs) will play an important role in future mobility systems, which promise a drastic reduction of traffic casualties, better road utilization and personalized mobility.
One of the main challenges to realize this future is to ensure that the presence of AVs in mixed traffic has a net positive effect on traffic safety. We posit that this would only be the case if their driving behavior and skills were indistinguishable from those of competent drivers, that is, if they show roadmanship.
This presentation outlines a methodology to operationalize roadmanship in such a way that it reflects the customs of a local driver population. It is based on assertions that describe the relationship among traffic participants. At any given time, a driver should maintain the value of applicable relationships within ``acceptable'' ranges for his/her driving to be deemed competent.
“Acceptability” is a subjective construct defined by expert opinion . Thus, the methodology includes a novel type of driving study that allows driver assessors to record their opinions about the quality of driving during driving sessions, link them to vehicle and traffic kinematics, and extract acceptable ranges for the assertions.
The resulting set of assertions can then be used to assess whether a driver, human or otherwise, is competent.
Speaker Biography
Arturo Tejada is a Senior Scientist at TNO’s Integrated Vehicle Safety department and part-time Assistant Professor of Safe Autonomous and Cooperative Vehicles at the Mechanical Engineering Department at Eindhoven University of Technology (TU/e). He holds bachelor’s degree in electrical engineering from the Pontificia Universidad Católica del Perú and MSc. and Ph. D. degrees in Electrical Engineering from Old Dominion University in Norfolk, Virginia (2006), where he specialized in hybrid system theory. He is the author of over 40 scientific publications in top peer-reviewed conferences and journals.
His work focuses on "teaching" self-driving vehicles how to interact with human drivers safely and socially. This is done by developing reference models of human driving from which design requirements for automated driving functions can be extracted and demonstrated. His work sits at the intersection of human factors, artificial intelligence, and motion control of automated vehicles. In addition, he supports OEMs and regulators in certifying and monitoring the safe behavior of automated vehicles in traffic.
Presentation File
Competent Driving Behavior
Category
Safety
Description