Mapping Route Optimization in Warehousing Environment Based on Improved Genetic Algorithm

Abstract

In large and structured warehousing environment, mapping route optimization based on mobile robot platform has been a challenging problem. To make the length of route shorter and form closed-loop earlier are the two goals of this optimization problem. Shorter route length could decrease the time and resource consumption during mapping. Earlier closed-loop could increase the accuracy of map and reduce closure detection errors. Because warehousing environment is structured, we could build an undirected graph in which the edges stand for passageways and the nodes represent the crossroads. With the two goals and the graph, we could transform the problem to a multi-objective Chinese postman problem. Genetic algorithm was used to solve this problem. We applied tournament selection operation and parthenogenesis. To improve the speed of evolution, the self-evolution process was introduced into the genetic algorithm. The results of experiments show the efficiency and validity of the proposed algorithm.