基于Simscape的APA-EPS系统建模及人机共驾转向控制策略研究

Simscape-Based APA-EPS System Modeling and Human-Machine Co-driving Steering Control Strategies Research

人机共驾是智能车辆实现高度自动驾驶之前的过渡阶段,由于驾驶员与驾驶辅助系统同时在环,存在驾驶权分配的问题,智能车辆的转向系统必须实时敏捷地执行驾驶员和驾驶辅助系统的转向控制指令。本文针对搭载于某智能MPV的平行轴式电动助力转向系统(APA-EPS)研究人机共驾转向控制策略。首先对APA-EPS系统进行Simscape建模,并通过台架试验验证了模型的准确度。随后基于模糊PID控制算法和模糊滑模控制算法分别设计了人驾和机驾模式下的控制策略。在驾驶员主导驾驶车辆时,根据距离影响度函数和角度影响度函数确定转向控制权重;在驾驶辅助系统主导驾驶车辆时,采用滑动时间窗口监测驾驶员的接管请求。仿真试验结果表明,人驾和机驾的控制策略均具有较好的转角跟随效果,人机共驾时转向控制权能够及时进行切换,且APA-EPS的转向盘转角响应迅速。

Human-machine co-driving is a transitional stage before intelligent vehicles achieving a high de-gree of autonomous driving.As the driver and the driving assistance system are in the loop at the same time,there is a problem of assigning driving rights for human-machine co-driving.The steering system of an intelligent vehicle must execute the steering control commands of the driver and driving assistance system in real time and with agility.The human-machine co-driving steering control strategies based on the parallel-axis electric power steering system(APA-EPS)in a smart MPV is studied in this paper.Firstly,the APA-EPS system is modeled in Simscape and the accuracy of the model is verified by bench tests.Subsequently,the steering control strategies for human-driving and machine-driving modes are designed based on fuzzy PID control algorithm and fuzzy sliding mode control algorithm respectively.When the driver is driving the vehicle dominantly,the steering control weight is determined by the dis-tance influence degree function and the angle influence degree function,when the driving assistance system is driv-ing the vehicle dominantly,the driver’s takeover request is monitored by sliding time windows.The simulation re-sults show that both the human-driving and machine-driving control strategies have good corner following effect,and in human-machine co-driving mode,the steering control rights can be switched in time,and the steering wheel an-gle response of the APA-EPS is rapid.