Genetic Algorithm-based Multi-robot Cooperative Exploration

Compared to single robot, multiple robots system has advantages for unknown environment exploration. The key problem is to allocate target points to multiple robots appropriately so that the multiple robots simultaneously explore different areas of the environment to guarantee minimum overall exploration time. However, the computation burden for optimal allocation exponentially increases with the number of robots and target points. Aiming at the problem, we propose a genetic algorithm-based coordinated multi-robot exploration algorithm on the basis of coordinated multi-robot exploration algorithm presented by Burgard. With its characteristics of random global searching and parallel computing, genetic algorithm is applied for allocating the target points to multiple robots. We describe that how the genetic algorithm can be applied to targets allocation to multiple robots. The technique has been tested extensively on simulation tests. The simulation results demonstrate that our method effectively distribute the target points to multiple robots over the environment. The multiple robots can accomplish exploration task quickly.