Global Path Planning of Mobile Robot for Agricultural Ground Robots

Abstract

Mobile robot navigation is a critical area of research, and global path planning stands as a pivotal technology within this domain. To address the challenges of slow convergence and potential local optimal solutions in global path planning, an improved ant colony algorithm is proposed. In the enhanced algorithm, the ant colony conducts purposeful searches by incorporating a heuristic function of terminal distance and direction, along with a step-by-step pseudo-random node transfer rule. This adaptive adjustment of target points strengthens the influence of the target point on the heuristic function, promoting more efficient path planning. Additionally, the pheromone updating strategy is improved, and a reward-punishment mechanism is introduced to enhance the ant colony's ability to recognize the optimal solution. By combining various pheromone allocation mechanisms, the convergence speed of the algorithm is significantly enhanced. Moreover, the improved algorithm considers the influence of obstacles on the path, ensuring both safety and the smooth attainment of the globally optimal path. In conclusion, the proposed improved ant colony algorithm offers a robust and efficient solution for the global path planning of mobile robots, particularly agricultural ground robots. By addressing the limitations of slow convergence and local optimal solutions, this research contributes significantly to enhancing the navigation capabilities of mobile robots in agricultural settings.