Abstract:When solving the problem of path planning of a mobile robot, defining constrains in environment modeling was hard, and was easily fallen into the local convergence by traditional genetic algorithm. By aimed at the problem above, and established the relationship among grid coordinates, grid number and grid state, the constrains definition of obstacles and effective path judging were simplified, at the same time, the cellular genetic algorithm with better diversity maintaining was also brought in, while optimizing the path by using fixed-length real number encoding. Finally, the simulation results show that the algorithm maintains the better diversity and improves the efficiency of the convergence because of the implicit mechanism of migration of the algorithm, which effectively solves the problem of path planning of the mobile robot.