Research On Multi-objective Optimization Algorithm Based On Improved Pelican Algorithm

Pelican optimization algorithm is a new stochastic nature-inspired optimization algorithm,which has strong mining ability to approach the optimal solution and the ability to explore the optimal area in the search space,and has a better performance in providing optimal solutions for optimization problems.In order to be able to apply it in the field of multi-objective optimization,this paper combines the idea of multi-objective optimization and improves the pelican optimization algorithm,and proposes a multi-objective optimization algorithm based on the improved pelican optimization algorithm(Multi-objective Optimization Algorithm Based on Improved Pelican Optimization Algorithm,MOIPOA),and applied it in the field of multi-UAV(Unmanned Aerial Vehicle,UAV)collaborative path planning.This paper conducts research around the above contents,and the main research contents are as follows:1.Aiming at the shortcomings of existing multi-objective optimization algorithms,such as easy to fall into local optimum,poor distribution uniformity of obtained non-dominated solution sets,and low speed and low search efficiency of pelican optimization algorithm itself in local search,a new algorithm,MOIPOA is proposed.Firstly,by combining multi-objective optimization ideas,external archive strategy,pareto optimal principle,non-dominated sorting algorithm and congestion operator,the applicable field of pelican optimization algorithm is extended to the field of multi-objective optimization;secondly,by using randomness the sobel sequence replaces the pseudo-random number generator to initialize the population,and further improves the uniformity of the initial population and the obtained non-dominated solution set distribution;finally,at the stage where the algorithm itself moves towards the prey,an adaptive dynamic factor is introduced to make the population.The individual in the algorithm realizes self-adaptive position update,which improves the convergence speed of the algorithm and performs better local search,and the effectiveness of the proposed algorithm is verified by experiments.2.Aiming at the safety of the UAV’s navigation and the coordination problem required by the mission,it is proposed that the space-time coordination test be used as one of the criteria for screening the UAV path planning scheme,and a multi-UAV cooperative path planning method based on MOIPOA is proposed.First of all,the path planning method of single UAV is studied,the evaluations criteria of the optimal path are discussed,and the path planning model of UAV is built,so that the engineering problem of UAV path planning can be transformed into a simple optimization problem;secondly,combined with the idea of multi-objective optimization,the single-UAV path planning model is used to transform the multi-UAV path planning problem into a multi-objective optimization problem for solution;finally,the space-time coordination test is added to the multi-UAV collaborative path planning based on MOIPOA.In the method,the feasibility and superiority of the multi-UAV collaborative path planning method based on MOIPOA are verified through experiments in multiple scenarios.