Research On The Application Of Robot Path Planning Based On Grouped Ant Colony Optimization Algorithm

Robot path planning has become an important part of the robot research field. It can guarantee that the robot completes the task safely and it is one of symbols of robot intellectualized degree. Ant Colony Optimization(ACO) algorithm which is proposed by Italian scholar Dorigo is a new heuristic optimization algorithm and it is inspired by analogy of behavior of real ants, when looking for foods. In the aspect of the complex optimized problems, the Ant Colony Optimization algorithm has become a newly competitive power solution algorithm which has a good development prospects.This paper mainly studies the robot path planning problem based on Ant Colony Optimization algorithm in the global static environment.Firstly, in-depth research on the basic principle and mathematical model of Ant Colony Optimization algorithm, including pheromone trail, heuristic, transition probability related knowledge and the solving problem of application to classical problem of TSP.Secondly, this paper studied on using Ant Colony algorithm to optimize the robot path planning, including the grids method. Through the analysis we have the conclusions:Ant Colony Optimization algorithm has many shortcomings such as: search speed is slow, easily to fall into local optimal and slow convergence speed.In order to overcome the disadvantages, this paper proposed the Grouped Ant Colony algorithm which has a better performance and applied it to solve the robot path planning problem. The Grouped Ant colony algorithm divided ants into two groups, put them at the starting point and goal point, respectively. Search path in opposite direction, it relieves pheromone superposition phenomenon, which contribute to the increased diversity of solution effectively. To further improve the performance of the Grouped Ant Colony Optimization algorithm, we propose Grouped Ant Colony Optimization algorithm based on several strategies, including the reset strategy, award-penalty strategy, the maximum-minimum strategy, target areas strategy, path crossover strategy. Simulation results show that the performance of the algorithm is better to optimize path in any complex environmentFinally, the paper summarizes the current research and points out the further future research as well.