APP下载

基于LQR算法的车道保持控制策略

2021-07-11施卫张晨

智能计算机与应用 2021年1期

施卫 张晨

摘 要:介绍一种基于LQR算法的车道保持控制方法。采用TLC与DLC联合预警模型结合驾驶员意图识别对车辆当前的行驶状态进行判断并在偏离时做出报警,当驾驶员未做出反应时车道保持系统对转向系统的输出转角值进行转向控制,帮助驾驶员纠正车辆偏离动作,在Carsim/Simulink环境下进行仿真,对比不同车速下横/航向偏差量与输出方向盘转角关系,结合车辆偏离预警算法实车测试,对该控制策略的可行性进行验证。结果表明,该控制策略的实时性和鲁棒性较好,具有实用价值。

关键词: 车道保持;车道偏离预警;驾驶员意图

文章编号: 2095-2163(2021)01-0020-05 中图分类号:U461.91 文献标志码:A

【Abstract】This paper introduces a lane keeping control method based on LQR algorithm. This paper uses TLC and DLC joint early warning model combined with driver's intention recognition to judge the current driving state of the vehicle and give an alarm when the driver deviates. When the driver does not respond, the lane keeping system controls the output angle value of the steering system to help the driver correct the vehicle deviation. The simulation is carried out in the CarSim / Simulink environment, and the transverse / heading at different speeds is compared. The relationship between the deviation and the output steering wheel angle, combined with the actual vehicle test of vehicle deviation warning algorithm, the feasibility of the control strategy is verified. The results show that the control strategy has good real-time performance and robustness, and has practical value.

【Key words】lane keeping; lane departure warning; driver's intention

0 引 言

随着全球化的日益推进,车辆智能辅助驾驶技术也在不断地发展,从最初的车身传感器对车辆行驶时路面动静态特征的监测捕捉,到后来的基于机器视觉和路面特征对车道线的检测识别[1],都充分说明了车辆智能控制技术的适应性与研究的必要性。

为了适应社会发展与市场需求,对车道偏离预警技术[2]和车道保持控制技术[3-4]的研发已然引起了越来越多研究人员和学者的兴趣与关注。本文在车道偏离预警及车道保持方面提出一种新的策略,通过建立基于路面附着系数TLC与DLC的联合预警模型,结合LQR算法的车道保持控制策略。相较传统预警及車道保持策略,该模型适用性较强且准确性较好。

1 车道偏离预警策略分析

本文所搭建的车道偏离预警模型为基于路面附着系数的TLC与DLC联合预警模型。……

登录APP查看全文