Multi-UAV cooperative path planning based on chaotic grey wolf optimiser

Abstract The success of cooperative unmanned aerial vehicle (UAV) missions relies on effective multi-UAV path planning. To address the issues of limited individual diversity and susceptibility to local optima during the population initialisation phase of the traditional grey wolf optimiser (GWO), this paper proposes an improved chaotic grey wolf optimiser (CGWO). The algorithm enhances population diversity by introducing chaotic initialisation to generate more uniformly distributed initial solutions. Combined with a chaotic local search strategy, it employs a dynamic balancing mechanism that allows individuals in the population to strike a balance between global exploration and local refinement, thereby breaking free from local optima constraints and accelerating optimal solution convergence. The algorithm was validated by using the CEC2017 benchmark functions and simulations of multi-UAV mission scenarios. The results clearly demonstrate that the improved algorithm significantly outperforms the original GWO and other common optimisation algorithms in terms of convergence accuracy and speed during benchmark testing. In multi-UAV mission scenarios, the enhanced algorithm excels in trajectory planning, flexibly avoiding obstacles while maintaining smooth flight paths for UAVs. Overall, this research provides a practical solution for coordinated multi-UAV operational path planning.