Urban Parking System Based on Dynamic Resource Allocation in an Era of Connected and Automated Vehicles

With the advent of automated vehicles, centralized online resource allocation offers opportunities in urban parking management to alleviate parking problem in the era of internet of things. In this paper, we propose a new parking system for large-scale resource management in an environment of mixed automated and human-driven vehicles. The system is formulated as a dynamic resource allocation (DRA) problem of mixed-integer linear programming (MILP) at each decision point with the objective of minimizing the total cost for users. Based on a tree representation of the solution space for the matching pairs, we combine Monte Carlo tree search (MCTS) and some heuristic rules to find a nearly global-optimal matching order (parking users and resources) within a short time. Simulation results show that this strategy can keep a good tradeoff between performance and computation efficiency, and this online resource allocation system can significantly alleviate urban parking problem in a mixed traffic environment.