Robot Path Planning for Human Search in Indoor Environments

Aiming at the problem of a mobile robot searching human in home environments, a gird model is built and a path planning method based on a modified genetic algorithm and an improved A* algorithm is proposed. First, the grid map is divided into several unit regions using Boustrophedon cellular decomposition. Then, a unit region planning method based on a genetic algorithm is applied to generate a region transition sequence, and an effective strategy to search every region is adjusted according to the robot's sensors. Meanwhile, the optimal path between two points is generated by an improved A* algorithm, so that the path is much shorter and the number of turns is greatly reduced. Finally, the simulation results verify that this method can provide an optimized path in known home environments effectively, based on that the robot can find human in the shortest possible time.