Research On Navigation Control System Of Rice Planter

Agricultural machinery has been listed in the key research field of "Made in China 2025".Auto-navigation technology is one of the trends in the development of agricultural machinery.Automatic navigation of agricultural machinery can effectively improve the working efficiency and quality,and reduce the labor intensity of farmers.However,several factors such as poor working conditions and high cost limit the application and popularization of intelligent agricultural machinery.In order to solve the problems of low degree of intelligence and control difficulty under poor working condition,the traditional rice planter is reformed,and complete coverage path planning and path tracking control algorithm are studied.The main contents are listed as follows:(1)Automatic transformation of traditional tractors is taken.In view of the low level of electrical control of traditional agricultural,the Yanmar VP6 E rice planter is used as the experimental platform.By adding a stepper motor and linear actuator for the steering system and throttle control system respectively,the automatic transformation is realized and the requirements for the automatic navigation system are satisfied.(2)The complete coverage path planning algorithm is studied.To obtain the best working direction in the work area,an optimization objective function with minimum turning area is proposed.For path planning in turning area,the fifth-order Bezier curve is used to fit the turn path.Considering the kinematics constraints of agricultural machinery,the nonlinear-optimization model is established with the goal of shortest length,and is finally solved by the differential evolutionary.(3)For the pure pursuit model,the path tracking system tends to oscillate when the vehicle speed is high.A stability optimization method based on fuzzy control is proposed.In this method,a kinematics model of the agricultural machine is first established by taking the steering system as a first-order model.Then,the stability condition of the pure pursuit model is analyzed when tracking a straight line.Based on this stability condition,velocity and lateral deviation input,a fuzzy control model is established to adjust the lookahead distance of the pure pursuit model in real time.The experimental results show that the method has a lower travel distance at the speed of 0.3m/s and lower overshoot at the high speed of 1.0m/s than the traditional method.It effectively improves the stability of the path tracking system.(4)For the problem of low precision of system control caused by disturbances such as sensor bias error,a path tracking method based on linear active disturbance rejection controller(LADRC)is proposed.Firstly,the kinematics of agricultural machinery is linearized.The tracking differentialize is designed to filter the lateral deviation and obtain the differential signal.A linear extended state observer is also designed to estimate the disturbances and the movement state.And the linear PD controller is compensated by the estimated value of disturbance.This method reduces the cost of the navigation system by only measuring the lateral deviation of the agricultural machine without the heading angle deviation.The experimental results show that the maximum mean error of this method is-0.23 cm.It proves that the steady-state error of the system is highly reduced.