Research On Multi-robot Path Planning Based On Artificial Bee Colony Algorithm

In recent years, multi-robot systems have attracted more and more attention because of their high reliability, strong robustness and high efficiency. Path planning is a key technology of multi-robot systems. Multi-robot path planning is a multi-objective optimization problem.This thesis introduces the research status of multi-robot path planning and artificial bee colony algorithm firstly. Then two artificial bee colony algorithms are proposed to solve different kinds of multi-robot path planning problems. One is the improved artificial bee colony algorithm which is based on vector weighting method. This algorithm can guarantee real-time path planning requirements. The other is the multi-objective artificial bee colony algorithm which is based on Pareto domination approach. This algorithm can optimize the multiple objectives of the paths simultaneously and reasonably. What’s more, this algorithm can provide multiple optimized paths with only one execution.This thesis proposes the improved artificial bee colony algorithm which is based on vector weighting method. This algorithm is optimized according to the real-time performance of robot path planning. Three key path performance indicators, length, smoothness and safety are taken into account. The initialization process and the food source evaluation strategy of the standard artificial bee colony algorithm are optimized in this algorithm. A new modeling method of the environment map is proposed at the same time to adapt to multi-robot path planning task. Simulation results show that the real-time performance of the algorithm and the effectiveness of the paths are improved greatly compared with standard algorithms. Experimental results show that the improved artificial bee colony algorithm can achieve multi-robot path planning.Considering the convenience of the decision-making of multi-robot systems, it is necessary to get a set of optimum paths whose performance indicators are all superior in the same time. The concept of non-dominated sorting and crowding distance is introduced to the artificial bee colony algorithm in this thesis. Then the calculation method of crowding distance is optimized and the improved multi-objective artificial bee colony algorithm is presented. The selection and elimination methods of food sources mechanism and the foraging mechanism of the bees are modified to improve the operational efficiency of the algorithm. Finally, the multi-objective artificial bee colony algorithm is optimized overall according to multi-robot path planning task. Simulation results show that the proposed multi-objective artificial bee colony algorithm can solve multi-robot path planning task efficiently, and its performance is better than other optimization algorithms. Experimental results show that the improved multi-objective artificial bee colony algorithm can plan a number of reasonable paths for each robot.