关键词: 蚁群算法, 启发函数, 信息素权重, 收敛速度

Abstract: In order to overcome the influence of traditional ant colony algorithm which is easy to fall into local optimum and slow convergence speed, a simulation model of robot experimental environment is established by using grid map. The ant colony algorithm is improved and applied to the robot path planning. Firstly, considering the directionality from the starting point of the path planning to the target point, the easy to fall into the local optimal solution in the early stage and the convergence speed of the ant, a heuristic function adding the bidirectional search direction mechanism and the proportional coefficient guiding factor is proposed to avoid the algorithm selecting the area that runs away from the end direction or the disadvantage of walking the loop. Secondly, according to the different times of different road segments being selected, different pheromone weights are set, the importance of different road segments is strengthened, and the convergence speed of the algorithm is accelerated. Finally, the algorithm simulation is carried out on the matlab software platform. The simulation results verify the effectiveness of the proposed method.

Key words: ant colony algorithm, heuristic function, pheromone weight, convergence speed