自动驾驶汽车变道的决策及预测控制算法研究
何倩，钟蕾，邱健斌，张子阳，潘俊霖，王艳红
（成都工业学院 汽车与交通学院，四川 宜宾 664000）
摘要：针对自动驾驶汽车的变道问题，提出一种基于规则和模型预测控制的自动驾驶汽车变道决策和控制算法。首先建立以后轴中心为原点的车辆运动学模型，并在模型中加入模型与测量的不确定性因素，使之更为接近现实场景。然后完成两车道车辆变道场景建模，基于预测碰撞时间的概念设计一套适用于车辆变道的决策规则，自动驾驶汽车能够通过感知周围车的状态信息实时进行变道判定；利用三次贝塞尔曲线完成车辆的变道路径规划，并使用模型预测控制方法对规划的路径进行轨迹跟踪，设计代价函数，将跟踪控制问题转化为最优控制问题。最后利用MATLAB对算法进行仿真。结果表明，所提变道算法的自动驾驶汽车不仅能够完成实时变道的任务，还具有一定的抗干扰能力。
关键词：自动驾驶汽车；变道决策；路径规划；模型预测控制；贝塞尔曲线
中图分类号：U463.6；TP13 文献标志码：A doi：10.3969/j.issn.1006-0316.2024.04.006
文章编号：1006-0316 (2024) 04-0034-08
Decision-Making and Predictive Control Algorithms for Lane Changing of Autonomous Vehicles
HE Qian，ZHONG Lei，QIU Jianbin，ZHANG Ziyang，PAN Junlin，WANG Yanhong
( School of Automobile and Transportation, Chengdu Technological University, Yibin 664000, China )
Abstract： This paper proposes a rule-based and model-based predictive control decision-making and control algorithm for lane-changing of autonomous vehicles. Firstly, a vehicle kinematics model with the center of the rear axle as the original point is established, and the uncertainty factors of model and measurement are added to the model to make it closer to the real scene. Then the modeling of two-lane vehicle lane changing scenario is completed, and a set of decision rules applicable to vehicle lane changing is designed based on predicting time to collision. The autonomous vehicle is able to make lane changing decision in real time by sensing the status information of the vehicles around. The lane-changing path planning of the vehicle is accomplished using cubic Bessel curves, and the trajectory tracking of the planned path is carried out with the model predictive control method. A cost function was designed to transform tracking control into optimum control. Finally, MATLAB was utilized to simulate the algorithm, and the results show that the autonomous vehicle with the proposed lane changing algorithm is not only capable of completing the task of real-time lane changing, but also has the anti-interference ability.
Key words：autonomous vehicle；lane change decision-making；path planning；model predictive control；Bessel curves
———————————————
收稿日期：2023-12-22
基金项目：成都工业学院校级重点科研项目（2023YB003）；成都工业学院重点校级项目（2022RW002）；成都工业学院校级一般科研项目（2023YB024）；成都工业学院校级一般科研项目（2023YB025）；成都工业学院校级一般科研项目（2023YB026）；成都工业学院校级一般科研项目（2023YB028）
作者简介：何倩（1996－），女，四川南充人，硕士研究生，助教，主要研究方向为自动驾驶汽车决策与控制算法，E-mail：243052028@qq.com。
（成都工业学院 汽车与交通学院，四川 宜宾 664000）
摘要：针对自动驾驶汽车的变道问题，提出一种基于规则和模型预测控制的自动驾驶汽车变道决策和控制算法。首先建立以后轴中心为原点的车辆运动学模型，并在模型中加入模型与测量的不确定性因素，使之更为接近现实场景。然后完成两车道车辆变道场景建模，基于预测碰撞时间的概念设计一套适用于车辆变道的决策规则，自动驾驶汽车能够通过感知周围车的状态信息实时进行变道判定；利用三次贝塞尔曲线完成车辆的变道路径规划，并使用模型预测控制方法对规划的路径进行轨迹跟踪，设计代价函数，将跟踪控制问题转化为最优控制问题。最后利用MATLAB对算法进行仿真。结果表明，所提变道算法的自动驾驶汽车不仅能够完成实时变道的任务，还具有一定的抗干扰能力。
关键词：自动驾驶汽车；变道决策；路径规划；模型预测控制；贝塞尔曲线
中图分类号：U463.6；TP13 文献标志码：A doi：10.3969/j.issn.1006-0316.2024.04.006
文章编号：1006-0316 (2024) 04-0034-08
Decision-Making and Predictive Control Algorithms for Lane Changing of Autonomous Vehicles
HE Qian，ZHONG Lei，QIU Jianbin，ZHANG Ziyang，PAN Junlin，WANG Yanhong
( School of Automobile and Transportation, Chengdu Technological University, Yibin 664000, China )
Abstract： This paper proposes a rule-based and model-based predictive control decision-making and control algorithm for lane-changing of autonomous vehicles. Firstly, a vehicle kinematics model with the center of the rear axle as the original point is established, and the uncertainty factors of model and measurement are added to the model to make it closer to the real scene. Then the modeling of two-lane vehicle lane changing scenario is completed, and a set of decision rules applicable to vehicle lane changing is designed based on predicting time to collision. The autonomous vehicle is able to make lane changing decision in real time by sensing the status information of the vehicles around. The lane-changing path planning of the vehicle is accomplished using cubic Bessel curves, and the trajectory tracking of the planned path is carried out with the model predictive control method. A cost function was designed to transform tracking control into optimum control. Finally, MATLAB was utilized to simulate the algorithm, and the results show that the autonomous vehicle with the proposed lane changing algorithm is not only capable of completing the task of real-time lane changing, but also has the anti-interference ability.
Key words：autonomous vehicle；lane change decision-making；path planning；model predictive control；Bessel curves
———————————————
收稿日期：2023-12-22
基金项目：成都工业学院校级重点科研项目（2023YB003）；成都工业学院重点校级项目（2022RW002）；成都工业学院校级一般科研项目（2023YB024）；成都工业学院校级一般科研项目（2023YB025）；成都工业学院校级一般科研项目（2023YB026）；成都工业学院校级一般科研项目（2023YB028）
作者简介：何倩（1996－），女，四川南充人，硕士研究生，助教，主要研究方向为自动驾驶汽车决策与控制算法，E-mail：243052028@qq.com。
- 上一篇：高速列车转向架变型设计知识推送系统构建与应用
- 下一篇：钢轨快速打磨车动力学仿真研究
设为首页 | 加入收藏 | 免责条款
《机械》杂志版权所有 Copyright©2008-2012 Jixiezazhi.com All Rights Reserved
电话：028-85925070 传真：028-85925073 E-mail：jixie@vip.163.com
地址：四川省成都锦江工业开发区墨香路48号 邮编：610063
Powered by PageAdmin CMS