Adaptive Ant Colony Algorithm Based on Global Scanning

In recent years, with the widespread application of industrial robots, intelligent path planning algorithms have attracted much attention due to their advantages over traditional algorithms. Among them, the ant colony algorithm is studied due to its good robustness and faster convergence speed. For traditional ant colony algorithms, slow convergence speed and easily falling into local optima are two disadvantages. Hence, some strategies are introduced to improve its performance. They are distance factor direction selection rules, global scanning, adaptive weighting coefficients adjustment, and variable initial pheromone distribution. The strategies reduce the limitation of step size to some extent, and show significant advantage in grids with uneven obstacle distribution. By adopting the strategy of directional angles, the confusion in path finding caused by the increased reachable points is avoided. Experiments show that the algorithm has advantages of high efficiency and low complexity in solving large and complex maps, it greatly improves the speed of initial solutions construction and convergence.