UAV 3D Path Planning Based on Multi-Population Ensemble Differential Evolution

Abstract

The three-dimensional (3D) environmental path planning of unmanned aerial vehicles (UAVs) is studied systemically. The model has been constructed, including 3D environment, threat, path length, yaw angle, climb angle and height. All the objective costs have been converted into a single objective optimization problem by the linear weight method. The multi-population ensemble differential evolution (MPEDE) algorithm has been applied to the 3D UAV path planning, and has shown the best performance compared with the self-adaptive differential evolution (SaDE) and differential evolution (DE) algorithms. The MPEDE algorithm is feasible in solve the UAV path planning problem.