A target vehicle with a self-developed chassis is affected by the manufacturing process,the clearance of its steering mechanism is inevitable,the virtual position of its steering wheel is too large,making it impossible to ac-curately control front wheels to drive at the expected angle,and the tracking trajectory is characterized as an"S"shape.In order to address the impact of steering clearance on steering control,this paper proposes a method to opti-mize the impact of steering clearance on steering angle control based on deep learning,which selects a three-layer neural network as a carrier in combination with regression thinking,and carries out theoretical calculations based on the two degrees of freedom model of the vehicle.Finally,actual vehicle verification shows that the control param-eters optimized by deep learning can effectively reduce the impact of steering clearance on steering angle control.
关键词
靶车/转向间隙/深度学习/神经网络/转角控制
Key words
Target vehicle/Steering clearance/Deep learning/Neural network/Steering angle control
维普
万方数据