Social-Aware Robot Navigation in Urban Environmnets

Gonzalo Ferrer, Anaís Garrell and Alberto Sanfeliu

Abstract—In this paper we present a novel robot navigation approach based on the so-called Social Force Model (SFM). First, we construct a graph map with a set of destinations that completely describe the navigation environment. Second, we propose a robot navigation algorithm, called social-aware navigation, which is mainly driven by the social-forces centered at the robot. Third, we use a MCMC Metropolis-Hastings algorithm in order to learn the parameters values of the method. Finally, the validation of the model is accomplished throughout an extensive set of simulations and real-life experiments.


Videos—The following videos show the performance of our robot navigation approach:

Simulation: In order to optimize the parameters of the social-aware navigation model, a MCMC technique is carried out. Here is depicted an example of one experiment in a crowded environment.

Real Experiments: the environment corresponds to the Barcelona Robot Lab, and the the video depicts some navigation experiments in a real situation.

_