A path planning method for human tracking agents using variable-term prediction based on dynamic k-nearest neighbor algorithm

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, paths of agents are planned based on similarity between the predicted positions of humans and agents´ field of views, and the path length in the path planning is determined according to the consistency between the current prediction and the previous prediction of the future human positions. We conducted computer simulations and results showed that our path planning method works well for trajectories of human in a real environment.