Multi-UAV collaborative task allocation method based on LGMPA algorithm

The multi-UAV task allocation problem is a multi-constrained combined optimization problem. To obtain the best benefit of the problem, this paper establishes a task allocation model that matches the battlefield environment. And the path planning between UAV and target is completed simultaneously. The real flight distance is used to replace the straight-line distance to realize the tight coupling of task allocation and path planning. A modified marine predator algorithm is proposed firstly in this paper to better solve the task allocation model. The exploitation and exploration ability of the algorithm is enhanced by using a logarithmic spiral strategy and a Gaussian distribution estimation strategy. The improved algorithm is then discretized and applied to solve the multi-UAV task allocation model. The simulation results show that the use of the real flight distance instead of the straight-line distance can plan a more reasonable task execution sequence and get a better combat profit under the consideration of the real terrain and various types of threats. The improved algorithm can solve the model stably and effectively, with better solution accuracy and faster convergence speed.