Modeling Vision Utility for Side-by-Side Navigation of Robot-Human Pairs

Abstract

Side-by-side robot navigation has significant direct benefits; especially when robotic wheelchairs are used. In addition to navigation issues faced by any mobile robot, side-by-side navigation has some other challenges as well. Maintaining side-by-side formation, avoiding collisions with minimal disturbance to the side-by-side formation, and maintaining the optimal social relationship are some of those challenges. This paper presents a novel decision making model for human-robot side by side navigation. The development of the model was based on observations of real-world human behaviour and data collections done through a user study. The developed model was calibrated and tested using a simulator as-well-as a laboratory experiment.