Path Tracking Control for Autonomous Harvesting Robots Based on Improved Double Arc Path Planning Algorithm

Abstract

Focusing on the problems of big overshoot and long convergence time due to the large initial heading error, a new path tracking control strategy for autonomous harvesting robots based on improved double arc path planning is designed. Firstly, the improved double arc path planning algorithm includes global path planning and local planning. The shortest path distance is taken as the goal, so the optimal tangent arc path from the initial point to the reference straight line can be automatically planned by the global path planning. Due to the problems such as control delay in harvesting robot, the global path cannot be effectively tracked. By analyzing the influence of the non-mutable steering angle and control delay on path tracking, a new local path planning algorithm is proposed. Then, the two preview points obtained in global planning can be optimized dynamically, and the actual reliability of the algorithm can be enhanced. Secondly, the error compensation model is designed to compensate the installation error of the positioning antenna in real time, to satisfy the simplified condition of the two-wheeled vehicle model. Finally, by combining the large angle steering control method and the pure pursuit algorithm, a new hybrid control strategy is designed to solve the steering angle, and the path tracking function of harvesting robots is realized. Experiment results demonstrate that the hybrid control strategy can reduce the oscillation of driving path, enhance the convergence, then, improve working quality and efficiency of robots.