Investigation of the Model for Path Selection Behavior in Open Space

摘要

In this study, we investigate the model for the human strategy of path selection behavior in an open space. In the experiment, a moving obstacle was displayed in an immersive virtual reality system. Subjects were required to walk from a start point to a goal point in the virtual space and their walking trajectories were measured. The model for the path selection strategy, composed of two stages of global path planning and local direction control, was proposed to account for the experimental results. In the global path planning stage, the model determined a transit position that minimizes an equation consisting of terms about walking distance and dangerousness. In the local direction control stage, a walking direction was calculated at every step to avoid a moving obstacle. The results of simulation well accounted for the experimental results in general. Moreover, we applied the model to the results of another experiment in order to investigate the generality of the model.