Research on AGV path planning based on improved artificial bee colony algorithm

When a traditional Artificial Bee Colony (ABC) algorithm is used to study an Automated Guided Vehicle (AGV) path planning problem, the algorithm suffers from slow convergence and low search efficiency and easily falls into local optimal. These problems with using ABC algorithms are because of the shortage of them in the processing of adaptation degree information sharing. To address these problems, an Improved Artificial Bee Colony (IABC) algorithm was proposed. Firstly, chaotic mapping population initialization was introduced to accelerate the convergence of the algorithm while improving the algorithm’s ability at local exploitation and global exploration. Secondly, Euclidean distance was introduced into the search for the detection bee to determine the effective neighborhood and the optimal solution within the neighborhood selected to generate new honey sources. Finally, the IABC algorithm was used for AGV path planning. The simulation comparison experiments of AGV path planning based on ABC and IABC algorithms were conducted in a raster map, and the simulation results showed that the IABC algorithm had a faster convergence speed and better local search capability.