Authors

Abstract

Path planning is crucial for agricultural machinery navigation. To address the issue of operational path planning in fields with obstacles, this paper proposes a method for obstacle avoidance path planning in farmland by combining an improved whale optimization algorithm with Dijkstra's algorithm. The population initialization is conducted using Tent mapping and a nonlinear convergence factor α^* is introduced to reduce the oscillation and instability of the traditional whale optimization algorithm. By using the grid method to model the environment of the target field, the field is divided into multiple regular subplots. The improved whale optimization algorithm is employed to determine the optimal traversal order of these subplots. Subsequently, Dijkstra's algorithm is applied to find the shortest path connecting the subplots, achieving global obstacle avoidance path planning for farmland. Taking a rectangular plot of land in Jiaolai Town, Jiaozhou City, Qingdao as the target area for this study, the experimental results indicate that this method achieves a coverage rate of 100% in the plot coverage path experiment. Additionally, the path redundancy rate is 4.87%, which represents a reduction of 1.63% compared to traditional algorithms. This research method is applicable to regular plots, but it still has limitations for irregular plots or those with curved boundaries.

Abstract in Chinese

路径规划是农机导航的关键。针对地块中存在障碍物的作业路径规划问题，本文旨在提出一种基于改进鲸鱼优化算法与Dijkstra算法相结合的农田避障路径规划的方法。本文通过利用Tent映射进行种群初始化，以及引入非线性收敛因子α^*，降低传统鲸鱼优化算法的振荡性以及不稳定性。通过栅格法对目标地块进行环境建模，将地块分成多个规则子地块,通过改进鲸鱼算法求解子地块最佳遍历顺序，再利用Dijkstra算法进行子地块之间连接最短路径，实现农田全局避障路径规划。以青岛市胶州市胶莱镇的一块矩形地块为本次研究目标地块，实验结果表明：本方法在目标地块覆盖路径实验中，地块覆盖率达到100%，路径重复率为4.87%，路径重复率较传统算法减少1.63%。