UAV online path planning based on dynamic multiobjective evolutionary algorithm

Online path planning (OPP) is the basic issue of some complex mission and is indeed a dynamic multi-objective optimization problem (DMOP). In this paper, we use an OPP scheme in the sense of model predictive control (MPC) to continuously update the environmental information for the planner. For solving the DMOP involved in the MPC-like OPP a dynamic multi-objective evolutionary algorithm based on linkage and prediction (LP-DMOEA) is proposed. Within this algorithm we selectively collect the historic Pareto sets and construct several time series to present the changing tendency of the dynamic Pareto set so as to properly guide the search process. Besides, a posterior method is introduced to select executive solution from the output of the LP-DMOEA. Experimental results show the advantage of the LP-DMOEA over restart method on three benchmark problems. The effectiveness of LP-DMOEA based OPP algorithm is also validated by the simulation results of a simple military case.