Research On Multi-objective Genetic Algorithm For Multi-robot Path Planning

In view of the importance of multi-objective optimization technology in engineering, economy, management, military, and so on, the research on multi-objective optimization has been paid more attention, it has developed into a new branch of science and showed great vitality in application, but the techniques of multi-objective optimization for robot path planning research are few. Multi-robot path planning is a complex multi-objective optimization problem, and the genetic algorithm is a global optimization, auto-adapted, probability search algorithm, which uses the experience of biological natural selection and genetic mechanism for reference, owing to its unique superiority and robustness in solving the complex system optimization, it becomes a very effective method in solving multi-objective optimization problem.This paper dwells on the status quo of the research, basic principles, and representative algorithm on the multi-objective optimization, and the main task and the issues to be resolved of Multi-objective genetic algorithm. The paper presents two practical methods of the multi-objective optimization genetic algorithm suitable for the path planning of multi-robot:(1) One is multi-objective hybrid genetic algorithm. In the basis of setting up the multi-objective model of a multi-robot path planning problem, the algorithm Focus on description of interference evaluation between paths and information exchange; the algorithm introduces a uniform and a large-scale initialization method in population and heuristic insert, delete, repair and smooth operators, according to the characteristic of multi-robot path planning. The paper uses the adaptive optimization to solve the phenomenon of "precocious" in genetic algorithm, which can effectively optimize a number of performance indicators of the problem at the same time. In a complex working environment, Simulation results show that Pareto solutions of the proposed algorithm are more, more widely distributed, and more even.(2) The other is Multi-objective optimization genetic algorithm based on Category groups. Through a series of classification, elite individuals of the entire group, Pareto the best individual, are separated and gived a high fitness to increase their probability of being selected. The algorithm not only has a high convergence, but also has the solution set for high quality. Simulation results confirm the correctness and high efficiency of the algorithm.