Research On Control Strategy Of Tractor Driving Robot Autonomous Steering System

In recent years,with the rise of robot technology and intelligent control technology in the field of vehicle applications,wheeled tractors has already been seen as an important machine for agricultural operations,the development of intelligent tractor automatic driving technology is an inevitable trend of future agricultural development in China.At present,the researches on the application of robot technology to agricultural vehicles at home and abroad have not been open to the public,and the control technology of vehicle self-steering by driving robot is rarely reported.In this paper,according to the characteristics of the tractor operating environment,high operational requirements and its influence in the process of turning from the ground resistance torque,a set of driving robot autonomous steering system is designed based on the tractor driving platform.The main contents are as follows:The overall design of the tractor driving robot autonomous steering system was carried out.The mechanical structure of the autonomous steering system and the design ideas of the autonomous steering control system based on the torque information were introduced.Combined with the design principles and objectives of the whole system,the main elements in the control system components were compared and selected,while its composition and working principle were also elaborated.The relationship among the steering wheel angle,the front wheel angle and the heading angle of navigation was obtained by analyzing the kinematic characteristics of the tractor independent steering system.Meanwhile,the main components of the autonomous steering system were analyzed,and the dynamic model was established.Based on that,the simulation model of the steering control motor with the reduction ratio was set up in the MATLAB/Simulink environment.According to the requirements of the steering system,the four control algorithms of the steering motor based on the ground torque and other parameters are studied,and the independent steering of the tractor is realized.Based on the motor model,the traditional PID controller,the fuzzy control controller based on the torque,the RBF-PID controller based on the torque and the fuzzy neural PID controller based on the torque/corner are designed.According to the information of the steering torque of the system,the fuzzy rules and membership functions of the fuzzy controller are determined.The PID control parameters are trained by the RBF neural network,and thus the PID control parameters are adjusted by the RBF neural network,and the parameters of the PID controller are determined by the critical scale method.The fuzzy PID control is based on the torque/rotation information,and the self-learning ability of the neural network is used to train the fuzzy rules for the adaptive dynamic output PID control parameters.In the MATLAB/Simulink,the four strategies are used to simulate the control system.Through comparison and analysis,the fuzzy neural PID control has the better follow-ability,robustness and a strong adaptive ability to the autonomous steering control system.Altium Designer software was used to complete the design of the autonomous control system,and the autonomous steering system test platform was set up.The tractor driving robot was independently turned to the ground test on different road surfaces of the school automobile experiment center square to collect the test data.And the traditional PID,dynamic fitting PID control method and FNN-PID control method are used to track the steering wheel angle through the measurement and control system respectively.The conclusion is as follows:the maximum overshoot of the dynamic fitting PID control method is 7.08%,the adjustment time The maximum overshoot of the FNN-PID control method is 5.56%,the adjustment time is 1.24s,the response speed is fast and the overshoot is small,and the tracking performance is better than that of the traditional PID control overshoot The results show that the autonomous steering system based on the identification of torque information has the better tracking ability and stability,and can realize the tractor self-steering function better.