Welcome to Xiaomi Yang’s PhD Defense “Improving pre-crash safety systems: Using computational driver behavior models and efficient sampling methods.”
Xiaomi Yang is a PhD student in the Vehicle Safety division of the Department of Mechanics and Maritime Sciences at Chalmers University of Technology. She has been one of the Early-Stage Researchers (ESRs) in the Horizon 2020 EC project SHAPE-IT, an Innovative Training Network project under the Marie Skłodowska-Curie Action program, coordinated by Chalmers. Her research has focused on the safety impact assessment of automated vehicles using counterfactual simulations. This include the development of statistical methods to improve the efficiency of both data collection with human participants (predictive Bayesian optional stopping) and scenario generation using computational behavior models (active sampling), as well as aspects of AEB algorithms improvements using behavior data.
Venue: The defense will be held in room Vasa A at Chalmers Campus Johanneberg. The entry to the room is in the middle of the hall you enter from Vera Strandbergs Allé 8, 411 33, Gothenburg, Sweden.
Thesis: The thesis Kappa and links to included publications can be found here. A Teams link is available for people that cannot join us physically (see the bottom of the email for more details).
Opponent: Meng Wang - TU Dresden
The grading committee:
Assoc. Prof. Charlie Klauer; Virginia Tech and Virginia Tech Transportation Institute and Virginia Tech
Assoc. Prof. Arno Eichberger; TU Graz
Assoc. Prof. Anna Anund; Linköping’s Universitet and VTI
Assoc. Prof. Ola Benderius (backup); Chalmers M2/VEAS
Main supervisor and defense chair: Jonas Bärgman (Bitr. Prof.; Chalmers)
Co-supervisor: Carol Flannagan (Research Prof.; University of Michigan Transportation Research Institute, UMTRI)
Examiner: Marco Dozza (Prof.; Chalmers)
Connected to the SAFER partner-project SHAPE-IT.
The defense starts with a presentation by Xiaomi for ~25minutes. We then take a short brake, after which the opponent will ask questions, followed by questions from the grading committee, and finally some questions from the audience.
We will serve something to eat after the defense. No registration is needed.
Welcome!
PhD thesis abstract
In-vehicle pre-crash safety systems have been used by the automotive industry for some time as integral parts of advanced driver assistance systems (ADAS) and automated driving systems (ADS), in order to improve traffic safety worldwide. Among the methods used to assess these systems, virtual safety assessment methods have been shown to have great potential and efficiency. These methods are likely to continue to play a very important role in assessing vehicle safety at all levels of automation. The ultimate aim of this thesis is to enhance the safety performance of conflict and crash avoidance systems through the use of computational driver behavior models. The work first addresses this aim by incorporating behavior models into pre-crash safety systems; the second part focuses on overcoming methodological challenges encountered in safety system development and assessments.
