Xiaolong Tian, Jin Peng, and Nan Wang
Unmanned aerial vehicle (UAV), collaborative path planning,particle swarm optimisation (PSO), dynamic factor
Multiple police unmanned aerial vehicles (UAVs) arrive simultane- ously at a target site, reconnoitering from various directions to fully cover the area, which is crucial for public safety. Traditional particle swarm optimisation has fixed learning factors for particle history and global best positions, which hinders its adaptation to iterative changes. A multi-UAV path planning method based on an improved algorithm is proposed. It introduces a dynamic balance learning factor and adjusts with iterations to balance local and global search. A Cauchy Gauss hybrid mutation strategy also speeds convergence. Considering UAV constraints, it shortens average cooperative arrival time by 10.4% and improves convergence speed by 25% over traditional methods.
Important Links:
Go Back
