Cooperative driving method improves safety despite imperfect vehicle communication
SIGN-MAPPO: State-Intent-Aware Graph Multi-Agent Reinforcement Learning for Cooperative Driving under Imperfect V2X
Robotics
Summary
Handling traffic at intersections with mixed human-driven and automated cars is tricky because drivers don’t always share their plans, and car communication isn’t perfect. The authors propose a new computer method that uses a special graph and prediction to help automated cars decide when to stop or go safely. Their system also checks for conflicts before acting to avoid accidents even if some messages are wrong or delayed. In simulations, this method completed nearly all crossings safely and quickly, showing it stays reliable even with communication problems or unpredictable human drivers.
What this means in practice
- •For automotive engineers: Develop cooperative driving software for automated vehicles that perform safely and efficiently at mixed-traffic intersections despite imperfect V2X communication.
- •For intelligent transportation system designers: Incorporate state-intent-aware policies in traffic management systems to improve coordination among connected and human-driven vehicles in urban areas.
Authors
Zide Fan, Jiahui Yuan, Kaixin Zhang, Junhao Fan, Xiaohe Li, Yiru Wang
Abstract
Cooperative driving at mixed-traffic unsignalized intersections is challenged by imperfect vehicle-to-everything (V2X) communication, human-driven vehicle (HV) intentions that cannot be directly communicated, and the potential incompatibility of independently generated crossing decisions. This paper proposes State-Intent-Aware Graph Multi-Agent Reinforcement Learning for Cooperative Driving under Imperfect V2X (SIGN-MAPPO) to address these challenges through three coordinated mechanisms. Each decision-eligible connected and automated vehicle (CAV) first reconstructs a receiver-specific traffic graph from its ego state and the peer-state records available to it, from which a shared graph-attention policy generates a raw Stop/Go intention augmented with predicted HV-intent features. CAV intentions are then exchanged through timestamped vehicle-to-vehicle (V2V) messages, and a local conflict-aware arbitration layer converts each CAV's intention into a safe executable action despite potentially inconsistent peer information. In the principal Simulation of Urban MObility (SUMO) evaluation, SIGN-MAPPO achieves a 97.53% completion rate, a 0.10% collision rate, and a mean intersection traversal time of 35.28 s. Ablation studies further confirm the complementary contributions of HV-intent prediction and CAV-intent exchange to this performance. The trained policy also remains robust under the evaluated communication-delay and packet-loss ranges, and cross-scenario tests show complete traffic clearance across the tested settings, though safety margins narrow under compound unpredictable HV behavior.