Path tracking control of crawler tractor based on adaptive adjustment of lookahead distance using sparrow search algorithm

Abstract

The complex and changeable farmland environment brings great challenges to the path tracking and stable control of autonomous agricultural machinery. To improve the efficiency and stability of path tracking control for autonomous agricultural machinery before entering the path from the headland and after entering the path, a path tracking control method for crawler tractor is proposed which can adjust the lookahead distance adaptively in Pure Pursuit control by using Sparrow Search Algorithm. In the method, the lookahead distance is regarded as sparrow individuals, and then the optimal target point is selected by simulating natural selection. Here, two lookahead area delineation schemes are set in light of the current lateral deviation of the agricultural machinery, and the lookahead area is preprocessed before searching the optimal target point to improve the efficiency and performance of the algorithm. Then, the current position information is substituted into the tracking control model of crawler tractor, and the position at the next moment is predicted under different lookahead distances. The predicted position is compared with the preset path to obtain the axial deviation and the heading deviation at the next moment. Additionally, the fitness function is constructed with the prediction deviation as the variable, and the optimal target point in the lookahead area can be searched by comparing the fitness values. In order to verify the feasibility of the proposed method, the ‘'entry-line and straight-line' tracking test of agricultural machinery under different initial postures and the curve tracking test with time-varying curvature were carried out on the platform of tracked tractor. The test results show that the average value, maximum value and standard deviation of the lateral deviation and the heading deviation after the agricultural machinery reaches the stable tracking state are not more than 0.030 m, 0.106 m, 0.034 m, 0.275 °, 1.528 °, 0.321 ° respectively in the 'entry-line and straight-line' test, and 0.059 m, 0.076 m, 0.175 m, 0.951°, 1.306°, 5.760° respectively in the curve tracking test. These results indicate that the path tracking control algorithm proposed in this paper improves the accuracy and stability of the path tracking of the crawler tractor, and has high adaptability to the curve path with time-varying path curvature, which meets the navigation requirements of the crawler tractor in the field operation.