Multiple path planning for a group of mobile robots in a 3D environment using genetic algorithms

We present the development, simulation and testing of a new approach using genetic algorithms for planning optimum paths for a group of mobile robots to be moved from arbitrary starting positions to final a number of targets in a known multi-obstacle 3D environment. The factors considered for fording optimum paths for the group of mobile robots are the size and location of obstacles in the environment and the topographical elevations of the environment. First, a digital picture of the environment is transformed into a grid map by a graphic simulator. The obstacles are mapped according to their location, shape and size. The ground elevations are represented using a color-coding scheme. The resulting grid map of the environment contains information about initial positions of the robots, target positions, obstacle locations and ground elevation. Hence, the location and size of obstacles and altitudes of the elevation of the environment are presented in the map. The genetic algorithm modules takes information about the environment from the grid map and search for optimum paths to move a group of mobile robots to the specified targets.