《计算机应用》唯一官方网站 ›› 2025, Vol. 45 ›› Issue (11): 3674-3681.DOI: 10.11772/j.issn.1001-9081.2024111569

• 先进计算 • 上一篇    

基于改进Q学习的智能车辆多场景安全制动算法

周贤文(), 龙潇, 余欣磊, 张依恋   

  1. 上海海事大学 航运技术与控制工程交通运输行业重点实验室,上海 201306
  • 收稿日期:2024-11-07 修回日期:2025-05-12 接受日期:2025-05-16 发布日期:2025-05-19 出版日期:2025-11-10
  • 通讯作者: 周贤文
  • 作者简介:龙潇(1997—),男,宁夏中卫人,硕士研究生,主要研究方向:强化学习、智能车辆制动
    余欣磊(1998—),男,浙江衢州人,硕士,主要研究方向:强化学习、智能车辆制动
    张依恋(1987—),女,浙江宁波人,讲师,博士,主要研究方向:港口无人装备状态估计与网络化控制。
  • 基金资助:
    国家自然科学基金资助项目(62176150);上海市地方院校能力建设项目(20040501400)

Improved Q-learning-based algorithm for safe braking of intelligent vehicles in multiple scenarios

Xianwen ZHOU(), Xiao LONG, Xinlei YU, Yilian ZHANG   

  1. Key Laboratory of Marine Technology and Control Engineering,Shanghai Maritime University,Shanghai 201306,China
  • Received:2024-11-07 Revised:2025-05-12 Accepted:2025-05-16 Online:2025-05-19 Published:2025-11-10
  • Contact: Xianwen ZHOU
  • About author:LONG Xiao, born in 1997, M. S. candidate. His research interests include reinforcement learning, intelligent vehicle braking.
    YU Xinlei, born in 1998, M. S. His research interests include reinforcement learning, intelligent vehicle braking.
    ZHANG Yilian, born in 1987, Ph. D., lecturer. Her research interests include state estimation and networked control of port unmanned equipment.
  • Supported by:
    National Natural Science Foundation of China(62176150);Capacity Building Project of Local Colleges in Shanghai(20040501400)

摘要:

针对智能车辆在混合交通流下的行驶安全问题,提出一种基于改进Q学习的智能车辆多场景安全制动算法。首先,根据路面情况与车辆参数建立三车模型,并分别模拟制动、跟车与变道场景。其次,对训练数据进行线性规划,以确保智能车辆存在安全制动的可能;同时,设置奖励函数,引导智能体在保证安全制动的基础上控制中车与前车、后车之间的距离尽可能相等。最后,结合区间分块方法,使算法能处理连续状态空间问题。与传统Q学习算法在制动、跟车、变道行驶场景下进行仿真对比实验的结果表明,所提算法的安全率由76.02%提高到100.00%,总训练时间降低为传统算法的69%。可见,所提算法安全性更好、训练效率更高,而且在制动、跟车与变道场景中均能在保证安全的前提下控制中车与前车、后车之间的距离尽可能相等。

关键词: 智能汽车, 智能制动, Q学习, 线性规划

Abstract:

Addressing the safety issues of intelligent vehicles driving in mixed traffic flows, an improved Q-learning-based algorithm for safe braking of intelligent vehicles in multiple scenarios was proposed. Firstly, a three-vehicle model was established according to road conditions and vehicle parameters, with scenarios of braking, car-following, and lane-changing being simulated respectively. Secondly, linear programming was applied to the training data to ensure the possibility of safe braking for intelligent vehicles. At the same time, a reward function was designed to guide the agent to perform safe braking while striving to equalize the distances between the middle vehicle and both the preceding and following vehicles. Finally, an interval-block method was incorporated to handle continuous state space issues. Simulation and comparison experiments were conducted in braking, car-following, and lane-changing scenarios, the results show that compared with the traditional Q-learning algorithm, the proposed algorithm has the safety rate increased from 76.02% to 100.00%, and the total training time reduced to 69% of the traditional algorithm. It can be seen that the proposed algorithm has better safety and higher training efficiency, and can ensure safety while striving to equalize the distances between the middle vehicle and both the preceding and following vehicles in braking, car-following, and lane-changing scenarios.

Key words: intelligent vehicle, intelligent braking, Q-learning, linear programming

中图分类号: