Improved Mixed Discrete Particle Swarms based Multi-task Assignment for UAVs

Abstract

Aiming at the problem of multi-task distribution for Unmanned Aerial Vehicle (UAV) swarm, a new type of multi-task distribution model for UAVs is established in this paper, various constraints are considered, such as bomb load and damage loss. To solve the multi-task allocation problem, an improved mixed discrete particle swarm optimization algorithm (IM-DPSO) is proposed, a two-dimensional particle coding matrix with task priority is designed, the genetic variation rules with particle update strategies are combined, and then the inertia weight and learning factors are optimized to enhance the algorithm's ability of solving the problem. Simulation results show that the improved algorithm can better solve the problem of multi-task allocation for UAVs under the distribution model.