Coverage path planning for mobile robot based on genetic algorithm

Abstract

Environment modeling for mobile robot is built up by using Boustrophedon cell decomposition method, and each sub-region is set numbers and basis point based on the characteristics of modeling, and connectivity relations among all sub-regions are established. All sub-regions are encoded by genetic algorithm (GA), and information of basis points between the sub-regions and sub-regions inside are set up and also achieved by GA, the optimal coverage sequences are obtained with GA, and in each sub-region a partial coverage is realized in the form of reciprocating movement, then problem of complete coverage for mobile robot is changed into a traveling salesman problem (TSP). Finally, the relationships between parameters of GA and search abilities are deeply studied, then the best parameters of GA are obtained. Simulation results show the effectiveness of GA for mobile robot's coverage path planning.