Three-Dimensional Path Planning for Unmanned Aerial Vehicles Based on Simulated Annealing Improved Particle Swarm Algorithm

Particle Swarm Optimization (PSO) algorithm has been widely applied in various path planning problems due to its simplicity and ease of implementation. However, traditional PSO algorithm suffers from issues such as poor search accuracy and susceptibility to getting trapped in local optima. In this paper, we propose an improved PSO algorithm that combines the Simulated Annealing (SA) algorithm for 3D path planning of unmanned aerial vehicles (UAV s) in mountainous environments. The probability jump characteristic of the SA algorithm is utilized to overcome the deficiency of PSO algorithm in getting stuck in local optima. To achieve a more uniform distribution of initial particle positions, we introduce the Logistic chaotic model for particle initialization. Additionally, non-linear inertia weight coefficients and monotonically decreasing individual learning factors, as well as monotonically increasing social learning factors, are employed to balance the global search capability and local refinement capability of the PSO algorithm. Simulation results demonstrate that the proposed algorithm outperforms the standard PSO algorithm and the Simulated Annealing-Standard PSO algorithm in terms of convergence accuracy and iteration speed, resulting in shorter planned paths.