Enhanced grey wolf optimization algorithm for group decision making in unmanned clusters

Abstract

unmanned platform cluster collaborative work will be a common mode to complete related tasks. The uncertainty and diversity of environment and task types make group decision-making a key technology to support unmanned cluster collaborative work. Gray wolf group has strong cognitive and cooperative ability, and can quickly track and surround prey in complex environment. The intelligent behavior characteristics such as information cognition and division of labor and cooperation are consistent with the decision-making needs of unmanned group. As a new swarm intelligence optimization algorithm, grey wolf optimization (GWO) algorithm is very suitable for solving such complex decision optimization problems. Aiming at the shortcomings of traditional GWO algorithm, such as slow convergence speed and easy to fall into local optimum, an enhanced grey wolf optimization (EGWO) algorithm is proposed to decompose the cooperative hunting movement of wolves in stages. The mathematical model mapped from search and track, pursuit, encirclement and attack prey is updated by three mathematical methods: nonlinear time-varying weighted weight differential mutation disturbance factor Archimedes spiral. The task stage switching model is constructed to deal with the state mutation in real time to enhance the correct decision-making and rapid response of the cluster. The experimental results show that the egwo algorithm can effectively improve the optimization accuracy and convergence speed of group decision-making.