抄録
This paper deals with a multi-agent path planning problem for tracking humans in order to obtain detail information such like human behavior and characteristics. To achieve this, path of agents is planned based on a clustering method, that is, agents follow paths which minimize Kullback-Leibler (KL) divergence between the intensity of humans existing and the intensity of field of view of agents calculated from predicted human positions and planned paths of agents. When prediction of human movement is accurate, the long-term prediction of human position would improve the performance. Since the prediction is not always accurate, the number of prediction steps is determined according to the difference between the current prediction and the previous prediction of the future human position. We conducted computer simulation and results showed that our path planning method works well even under changing circumstances.
