Robot Path Planning Research Based On Ant Colony And Genetic Algorithm

In the field of intelligent robotics research, Path planning technology is a hotissue, searching the best path is the purpose of smart robot path planning, whichcannot bump into any obstacle between the starting point and the target point intheir working environment containing obstacles. Path planning methods includethe traditional path planning methods and artificial intelligence path planningmethods. Generally, intelligent path planning method not only has theapproximation linear, self-learning, self-organizing function, but also hasfault-tolerant ability, where the typical representatives are ant colony algorithmpath planning method and genetic algorithm path planning method.Ant colony algorithm (ACA) is an artificial intelligent algorithm, and thegenetic algorithm (GA) also belongs to intelligent algorithm, and both algorithmshave implicit parallel search capability. The ant colony algorithm can make fulluse of pheromone positive feedback information to accelerate convergence onoptimal problems, and has strong robustness; genetic algorithms is similar to thebiological genetic crossover, mutation operation, saving outstanding individualgenes during evolution and eventually converge to the optimal solution, howeverwhen the population too much, the cross operation will become very complicated,therefore, making full use of the advantages of the two algorithms to solve the pathplanning problem is a hot issue.The path planning algorithm in the paper makes full use of both the ACA andGA advantages, based on the ant colony algorithm that the Max-Min Ant System(MMAS). The path planning algorithm by using local optimum path informationand global optimal path information to update pheromone of the optimal paththough adding weight factor; and import the idea of dual ant colony andcross-factor, and come out better path by crossing the iterative optimal path andthe global optimal path, and update the global optimal path by the better path.