Research on UAV Path Planning Based on 3D Scene Modeling

In response to the problem of simplifying the simulation experimental environment and obtaining experimental results that cannot meet the actual navigation needs in the research of existing unmanned aerial vehicle (UAV) path planning algorithms, this study establishes a three-dimensional (3D) environment model of actual terrain based on digital elevation models (DEM), meanwhile uses functions to establish random 3D maps, and selects particle swarm optimization (PSO) algorithm and ant colony algorithm (ACO) to carry out the UAV path planning under circumstances of DEM and random 3D map, and the four sets of experimental results obtained are compared and analyzed. Compared to the DEM model, the random terrain has a lower complexity, allowing the fitness values of both the PSO and ACO algorithms to converge more quickly, with less time required to complete 100 iterations. For both algorithms, the PSO algorithm takes less time, only about one-third of the time taken by the ACO algorithm. It is recommended to use the DEM that derived from the real terrain environment instead of a simplified math-derived environment model in UAV path planning research.