Hierarchical Multi-UAV Path Planning for Urban Low Altitude Environments

Aiming at the demand of multi-UAV missions in urban low-altitude environment, combined with the concept of low-altitude economy, the urban environment is divided into multiple layers in the vertical direction to accomplish path planning. The traditional ant colony algorithm has problems such as slow planning speed and easy to fall into local optimization when performing UAV urban environment path planning. In order to improve these problems, we combine the Ant Colony Algorithm (ACO) with the Particle Swarm Algorithm (PSO), and utilize the early and fast convergence of the PSO to generate a suboptimal solution as the initial condition of the pheromone distribution of the ACO. Meanwhile, parameter adaptive optimization and forbidden strategy optimization were performed for the ACO, and parameter adaptive optimization and local optimum optimization were performed for the PSO. In addition, the energy consumption constraints and layering constraints of the UAVs are considered, and finally the paths are optimized using NURBS curves so that each UAV can reach the end point of its respective layer. The experimental results show that the average optimal fitness of the algorithm is improved by 22.2% and the algorithm running time is reduced by 33.0% compared to the traditional ACO.