Traffic expertise meets residual RL: Knowledge-informed model-based residual reinforcement learning for CAV trajectory control
交通专家与残差强化学习相遇:基于知识的模型驱动残差强化学习用于智能交通车辆轨迹控制
机构 * Department of Civil and Environmental Engineering, University of Wisconsin-Madison, Madison, WI, 53706, USA(土木与环境工程系,威斯尼大学麦迪逊分校)
专题命中 模型式强化学习 :environment model(abstract);model-based reinforcement learning(abstract);model-based RL(abstract);分类 cs.AI、cs.LG
AI总结 本文提出基于知识的模型驱动残差强化学习框架,用于智能交通车辆轨迹控制,结合交通专家知识与传统控制方法,提升学习效率与交通流平滑度。
Comments Accepted by Communications in Transportation Research
Journal ref Communications in Transportation Research 4 (2024): 100142