Coverage path planning for mobile robots using genetic algorithm with energy optimization

Abstract

Coverage path planning (CPP) is one of the current researches for mobile robots. This study presents a new approach for solving CPP. In this approach the coverage area is divided into small squares where the squares diagonal is the size of the robot tool. Four fundamental motions in a square are defined, which are straight, left turn, right turn and U-turn. A cost function is taken in which a fixed cost for each fundamental motion in a square is used and costs for all moves are summed up to get an approximated cost for a path. This function makes it possible to find the better path between two paths with equal repetitive visits. Furthermore genetic algorithm (GA) is used to find the best path to cover an area. Via crossover, selection and mutation GA improves current paths which leads to optimal, near optimal solutions. Simulation results are taken with reasonable areas.