Three-Dimensional Obstacle Avoidance Path Planning for Agricultural UAV Based on Improved Ant Colony Algorithm

Obstacle avoidance is crucial for unmanned aerial vehicles (UAVs) in agriculture to perform tasks such as crop monitoring, precision spraying and picking assistance. The three-dimensional (3D) nature of orchards, with issues such as a larger search space, diverse obstacle shapes, and higher requirements for environmental adaptability, poses challenges for UAV obstacle avoidance. In this paper, a 3D obstacle avoidance path planning method for agricultural UAVs based on improved ant colony optimization (ACO) algorithm is proposed. Firstly, A 3D orchard simulation environment is constructed using MATLAB. Secondly, based on the characteristics of the orchard environment with large ups and downs of the terrain and multiple obstacles, the fitness function and heuristic information in the ACO algorithm have been modified. Thirdly, a leader ant optimization (LAO) algorithm is proposed by introducing the bellwether theory to improve the ACO algorithm. The LAO algorithm has been comprehensively evaluated on path optimal solution, obstacle avoidance capability, convergence speed and computation time. The experimental results demonstrate that the performance of LAO algorithm is optimal compared to the traditional ACO algorithm, genetic algorithm (GA) and particle swarm optimization (PSO) algorithm. The proposed LAO algorithm is suitable for UAV obstacle avoidance in orchards.