Socially-Aware Robot Navigation: combining Risk Assessment and Social Conventions

This thesis proposes a risk-based navigation method including both the traditional notion of risk of collision and the notion of risk of disturbance. With the growing demand of personal assistance to mobility and mobile service robotics, robots and people must share the same physical spaces and follow the same social conventions. Robots must respect proximity constraints but also respect people interacting. For example, they may not break interaction between people talking, unless the robot task is to take part in the conversation. In this case, it must be able to join the group using a socially adapted behavior. The socially-aware navigation system proposed in this thesis integrates both an assessment of a risk of collision using predictive models of moving obstacles, and an assessment of accordance with social conventions. Human management of space (personal space, o-space, activity space...) inspired from sociology and social robotics literature is integrated, but also models of behavior that enable the robot to make medium-term prediction of the human positions. Simulation and experimental results on a robotic wheelchair validate the method by showing that our robot is able to navigate in a dynamic environment avoiding collisions with obstacles and people and, at the same time, minimizing discomfort in people by respecting spaces mentioned above.