Multi-objective optimization of dive trajectory for morphing unmanned aerial-underwater vehicle

Abstract

A well-planned dive trajectory is the key for the trans-media maneuver and flight control of morphing unmanned aerial-underwater vehicles (MUAUVs). The main focus of this paper is the application of a watch-based multi-objective grey wolf optimizer (MOGWO) to dive trajectory optimization problems. A multi-objective trajectory optimization model considering the change of sweep angle and static moment is developed and parameterized by the Gauss pseudospectral method, and a multi-objective nonlinear programming problem is constructed. To overcome the defect that, in the MOGWO, the rest wolves blindly follow the best wolves, the watch strategy is introduced, which provides the wolves with the ability of independent exploration. Subsequently, the watch-based MOGWO is employed to generate the Pareto front, which is compared with that obtained through other multi-objective techniques. The simulation results demonstrated that the proposed method is more reliable and can obtain more widely distributed non-dominated solutions, indicating that the watch-based MOGWO is effective and feasible in dealing with multi-objective trajectory optimization problems with complex constraints. In addition, comparative studies on optimal trajectories demonstrated that the maneuverability and gliding ability of the MUAUV are improved through cooperation between the angle-of-attack, bank angle, and sweep angle.