A Novel Spacing Control and Velocity Optimization Method for Electric Vehicle Platoons Based on Communication Proximal Policy Optimization

The growing global car population has led to more traffic accidents, road congestion, and energy consumption. Eco-CACC (ecological cooperative adaptive control) has proven effective in addressing these issues. However, current eco-CACC strategies for battery electric vehicles (BEVs) depend heavily on models, resulting in performance degradation with model inaccuracies, and often overlook battery life, acceleration limits, and string stability. To bridge the research gap, this paper proposes a model-free eco-CACC strategy for BEV platoons to enhance driving safety, passenger comfort, battery lifespan, and energy efficiency. The strategy includes a model-free velocity optimization method for the leading BEV and a multi-objective spacing controller for following BEVs. The spacing controller, based on communication proximal policy optimization, ensures safe, efficient spacing within the platoon while considering acceleration limits and string stability. The velocity optimization method employs a closed-loop cost feedback mechanism to optimize the platoon's reference velocity in real time. Simulations across various road scenarios show that our strategy not only ensures safety and comfort but also significantly reduces energy consumption and battery capacity loss compared to benchmarks. The proposed eco-CACC is real-time, model-free, and adaptable to different road profiles, offering substantial improvements over existing methods.