Multi-Vehicle Automated Driving as a Generalized Mixed-Integer Potential Game
More Info
expand_more
expand_more
Abstract
This paper considers the multi-vehicle automated driving coordination problem. We develop a distributed, hybrid decision-making framework for safe and efficient autonomous driving of selfish vehicles on multi-lane highways, where each dynamics is modeled as a mixed-logical–dynamical system. We formalize the coordination problem as a generalized mixed-integer potential game, seeking an equilibrium solution that generates almost individually optimal mixed-integer decisions, given the safety constraints. Finally, we embed the proposed best-response-based algorithms within the distributed open- and closed-loop control policies.
Files
MVAD_final_version.pdf
(pdf | 3.04 Mb)
08672171.pdf
(pdf | 1.55 Mb)
Download not available