Contention-Resolving Model Predictive Control for Coordinating Automated Vehicles at a Traffic Intersection
Our earlier work established a contentionresolving model predictive control (MPC) framework for codesigning priorities and speeds for automated vehicles at an intersection. In this paper, we present an improved contentionresolving MPC design. We propose a new branch cost formulation for the decision tree that is constructed by contentionresolving MPC to handle the case where a vehicle is delayed multiple times before being allowed access to the intersection. All possible priority combinations are dynamically generated when constructing the decision tree. Based on the priority assignments, we design a decentralized control law to control vehicle speeds, which we show enjoys optimality properties under a specific priority assignment. We verify the effectiveness of our method through a simulation and a comparison with the First-Come-First-Serve scheduling strategy.
Publication Source (Journal or Book title)
Proceedings of the IEEE Conference on Decision and Control
Yao, N., Malisoff, M., & Zhang, F. (2019). Contention-Resolving Model Predictive Control for Coordinating Automated Vehicles at a Traffic Intersection. Proceedings of the IEEE Conference on Decision and Control, 2019-December, 2233-2238. https://doi.org/10.1109/CDC40024.2019.9029542