Abstract
Ant colony algorithm is mostly applied to enterprise, underwater detection, logistics optimization, is to follow the ant population based on spatial consciousness planning model, relative to two-dimensional path planning has more obstacles, more mixed space, more information storage, more complex operation logic and difficult to direct global planning, in the dynamic environment, put forward the global path planning and local path planning. Based on static knowledge of static environment, I propose an ant colony algorithm. On this basis, modify the ant path search strategy to jump out of some special shape obstacles; establish the probabilistic selection formula of the basic ant colony algorithm, add the target node information to the algorithm convergence; select the formula, and add the volatile coefficient to improve the adaptability of the algorithm. In the planning process, by analyzing the distribution of obstacles, the corresponding avoidance strategies are proposed for the different positions of obstacles. Simulation experiments show that the proposed algorithm can effectively realize the robot obstacle avoidance target in a complex 3D environment, and obtain the touch-free optimal or suboptimal path.

This publication has 10 references indexed in Scilit: