A Value Based Parallel Update MCTS Method for Multi-Agent Cooperative Decision Making of Connected and Automated Vehicles

2024-09-20

arXiv.org

AI Summary

To address the joint lateral and longitudinal decision-making challenge for connected and autonomous vehicles (CAVs) in multi-vehicle cooperative scenarios, this paper proposes a value-driven parallel-update Monte Carlo Tree Search (MCTS) method, formulated within a finite-horizon discounted-time multi-agent Markov game framework. We innovatively design a partially steady traffic-flow-guided parallel action analysis mechanism to enable efficient action pruning and synergistic depth-width optimization. Compared with conventional MCTS, state-of-the-art reinforcement learning, and heuristic approaches, our method significantly improves hazardous maneuver detection efficiency and policy robustness under large-scale stochastic traffic flows. Specifically, it increases throughput in cooperative zones by 12.7% and reduces collision rates by 63%. Both safety and efficiency metrics surpass those of human drivers, demonstrating superior comprehensive performance in real-world CAV coordination tasks.

Technology Category

• Multiagent Systems: Coordination and Collaboration
• Humans and AI: Planning and Decision Support for Human-Machine Teams
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

To solve the problem of lateral and logitudinal joint decision-making of multi-vehicle cooperative driving for connected and automated vehicles (CAVs), this paper proposes a Monte Carlo tree search (MCTS) method with parallel update for multi-agent Markov game with limited horizon and time discounted setting. By analyzing the parallel actions in the multi-vehicle joint action space in the partial-steady-state traffic flow, the parallel update method can quickly exclude potential dangerous actions, thereby increasing the search depth without sacrificing the search breadth. The proposed method is tested in a large number of randomly generated traffic flow. The experiment results show that the algorithm has good robustness and better performance than the SOTA reinforcement learning algorithms and heuristic methods. The vehicle driving strategy using the proposed algorithm shows rationality beyond human drivers, and has advantages in traffic efficiency and safety in the coordinating zone.

Problem

• Solves lateral-longitudinal joint decision-making for multi-vehicle CAVs
• Enhances MCTS with parallel updates for multi-agent Markov games
• Improves traffic efficiency and safety in cooperative driving scenarios

Innovation

• Parallel update MCTS for multi-agent CAVs
• Excludes dangerous actions efficiently
• Enhances search depth and breadth