A multiagent approach to autonomous intersection management
Journal of Artificial Intelligence Research
Efficient algorithms for collision avoidance at intersections
Proceedings of the 15th ACM international conference on Hybrid Systems: Computation and Control
Automatica (Journal of IFAC)
Hi-index | 0.00 |
Even with recent advances in electronics, sensors, and communication technologies, it is still a grand challenge to develop an autonomous transportation system in which each vehicle can autonomously coordinate with other vehicles and drive itself on a road with a safety guarantee. We propose an approach to address this challenge. We formulate a Model Predictive Control (MPC) problem to generate a feasible trajectory for a vehicle. Constraints are particularly designed to ensure inter-vehicle safety. To further achieve system-wide safety and liveness, we additionally propose several simple yet effective rules for vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) coordination. We establish system-wide safety and also liveness of the resulting autonomous traffic in various traffic situations such as single lane traffic, multi-lane traffic, and intersection crossing traffic.