A model for determining natural pathways for side-by-side companion robots in passing pedestrian flows using dynamic density

Abstract

Passing pedestrians is a common task for pairs while moving. Whilst pairs generally prefer Side-by-side walking mode, that mode tends to occupy more space in the pathway and reduces space for pedestrians traveling in the opposite direction than Leader-Follower mode in which one follows the other. Thus, humans often intuitively consider solutions to optimize the balance between side-by-side walking mode and moving space for others in passing. This is also a problem that designers of companion robots often have to solve. By discovering, modeling, and incorporating a new factor - the habit of moving with the flow and density in moving (called dynamic density) - this work proposes a novel model to determine natural navigation pathways for companion robot to pass multiple pedestrians walking in the opposite directions, mimicking human passing behaviors by taking into account this factor. Based on two experimental observations and data collections, the model was developed and then validated by comparing the pathways generated by the model and the natural moving plans of the pairs in the same situations. The simulation results show that the new model is able to determine moving plans of pairs in passing situations, similar to real decisions of humans.