A Routing-based Strategy to Socially Approach Multiple Individuals in Cluttered Environments

The use of autonomous mobile robots is expected to continue increasing in different parts of society in the next years. In this context, when not restricted to industrial environments only, it is essential to provide such devices with the ability to behave in a so-called socially acceptable manner. In the Human-Robot Interaction (HRI) field, a fundamental problem is to approach people, i.e., try to initiate and maintain an interaction. This task involves different challenges since some implicit nonverbal social rules must be respected, especially those concerning people's personal spaces (called proxemics). Considering the presence of multiple individuals to be approached, a combinatorial optimization aspect must also be tackled, where an optimal sequence of visits must be selected. In this paper, we formulate this problem as a Generalized Travelling Salesman Problem (GTSP) and combine it with a Probabilistic Roadmap (PRM) planner in order to account for obstacles in the environment. Numerical simulations show that the methodology produces length-optimized paths that are feasible, safe, and respect the social conventions.