基于模拟退火算法的无人驾驶车辆轨迹跟踪控制  

Research on Trajectory Tracking Control of Self-Driving Vehicle Based on Simulated Annealing Algorithm

作　　者：王浩[1] 林棻[1] 张尧文 

机构地区：[1]南京航空航天大学能源与动力学院车辆工程系,南京210016

出　　处：《重庆理工大学学报（自然科学）》2015年第11期106-111,119,共7页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(10902049);江苏省博士后科研资助计划项目(1302020C);中央高校基本科研业务费专项资金资助项目(NS2015015);南京航空航天大学研究生创新基地(实验室)开放基金资助项目(kfjj201404)

摘　　要：研究了基于模拟退火算法(SA)的无人驾驶车辆横向控制策略。建立了线性离散车辆运动学预测模型,考虑位姿偏差和控制量的变化设计目标函数,运用模拟退火算法(SA)对目标函数进行求解并获得当前时刻的控制增量,进而实现无人驾驶车辆的轨迹跟踪控制。通过在低速组合轨迹工况下的仿真结果表明:在不同的车速下,该控制策略能够保证车辆快速平稳地跟踪给定参考轨迹,位置误差能够控制在0.05 m之内,具有较好的控制效果。Lateral control based on simulated annealing（ SA） algorithm was investigated for trajectory tracking of self-driving vehicle. Firstly,the linear discrete kinematics model was established. Based on this model,considering position and pose error,the objective function and constraint condition were designed,and simulated annealing（ SA） algorithm was used for optimization problems and obtaining the optimal control variables and the trajectory tracking control was achieved. In the condition of low speed and combined trajectory,the simulation results show that in different speeds,the control strategy guarantees that the vehicle can follow the reference trajectory stably and rapidly with good effect,and the position error can be kept within 0. 05 m.

关 键 词：无人驾驶车辆 模型预测控制 模拟退火算法 轨迹跟踪 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]