| With the continuous aggravation of our aging population,the drain of rural labor force accelerates,which causes agricultural production cost to rise,the efficiency and competitiveness of agricultural production to decline constantly.In addition,the accelerated urbanization process also makes more and more farmers leave the countryside,resulting in a shortage of agricultural labor force,agricultural production power weakened.Intelligent agricultural machine equipment can improve operation efficiency,operation quality,land utilization rate,can solve the current domestic agricultural development predicament,in line with our agricultural development conditions.Path tracking is the key technology to realize intelligent agricultural machinery equipment.Therefore,based on model predictive control algorithm(MPC),path tracking control of agricultural machinery is studied in this paper.Firstly,the kinematic model of agricultural machinery was established based on the kinematic characteristics,and the two-degree-of-freedom dynamic model of agricultural machinery was established according to the force analysis.The kinematic model and dynamic model were validated by CarSim/Simulink,which provides the basis for the research of path tracking algorithm.Then,a fusion control algorithm based on MPC and Stanley is proposed to solve the problem that MPC computation is complex and sometimes the rappiness cannot be guaranteed.Based on the nonlinear state space model of agricultural machinery kinematics and nonlinear state space model of dynamics,the linear time-varying state space model was obtained by local linearization method,and the path tracking control law was designed by MPC algorithm.Based on the MPC algorithm,the weight coefficient and Stanley algorithm were fused.The lateral error and course error were taken as the input,and the weight coefficient was taken as the output.A fuzzy controller was established to dynamically adjust the weight coefficient.The joint simulation platform of CarSim and Simulink was built to track the expected path at different speeds.Simulation results show that the control performance of the proposed algorithm is better than that of the conventional MPC algorithm.The fusion algorithm is faster in the first half of the tracking process and smoother in the second half of the tracking process.The fusion algorithm uses 20 km/h,30 km/h and 40 km/h speeds to track the double-shift path with an initial deviation of 5 m.The on-line distances of the fusion algorithm are 15.8 m,16.2 m and 25.3 m,respectively,while the on-line distances of the conventional MPC algorithm are 17.46 m,18.9 m and 27.19 m.The linear tracking error is within ±2.5cm.Especially suitable for agricultural machinery initial position deviation and high speed applications.Finally,the hardware equipment was selected and the automatic driving system was designed.The hardware equipment mainly included the satellite navigation module and the steering control module.Real vehicle experiments were conducted on the linear path and U-shaped curve path respectively.The initial error of agricultural machinery tracking is 1 m straight line,the on-line distance of fusion algorithm is3.5m,and the on-line time is 1.75 s.After on-line,the average value of the absolute transverse error of agricultural machinery is 2.15 cm,and the maximum transverse error is 4.5cm,which meets the actual working requirements of agricultural machinery.At the same time,compared with the conventional MPC algorithm,the fusion algorithm designed in this paper has significantly improved the speed while ensuring the control accuracy. |
PDF Full Text Request