无人驾驶汽车转向系统参数自整定及优化研究

Research on Auto-Tuning and Optimization of Steering System Parameters for Autonomous Vehicles

针对已有的PID控制器在伺服系统中适应性差、参数调整困难的问题,对永磁同步电机伺服系统进行研究。提出了一种综合模型和规则的参数自整定策略,基于典型系统对三闭环PID控制器整定得到初值,其中利用模型参考自适应算法对电机进行辨识以得到速度环参数初值,采用共轭梯度法在参数初值邻域内寻优,进行Matlab/Simulink仿真分析,并通过了试验验证。结果表明,参数自整定下的控制器比传统控制器位置超调量上升了5%以内,上升时间下降了20%~50%。通过仿真与试验验证,参数自整定算法可以更优化地适应各种工况。

To overcome the difficulties of parameter tuning and to improve the adaptability of existing PID controllers in the servo system,the PMSM servo system is studied.A comprehensive strategy combining modeling with rules was proposed for parameter auto-tuning.Initially the three closed-loop PID controller was tuned to obtain the initial values based on a typical system.The model reference adaptive algorithm was used to identify the motor to obtain the initial values of the speed loop parameters.The conjugate gradient method was applied for optimization in the neighborhood of the initial values,and finally the Matlab/Simulink simulation was carried out to verify the experimental results.The results show that compared with the traditional controller,the position controller overshoot has increased by less than 5%with parameter auto-tuning,and the rise time has decreased by 20%~50%.The simulation and experimental results verify that the proposed parameter auto-tuning algorithm can be more optimally adapted to various conditions.