Towards S-NAMO: Socially-aware Navigation Among Movable Obstacles

TL;DR

This paper analyzes Socially-aware Navigation Among Movable Obstacles (NAMO), proposing new algorithms for social acceptability, developing a simulator for testing social obstacle handling, and conducting preliminary robot experiments.

Contribution

It introduces the concept of social acceptability into NAMO, develops a simulator for social obstacle evaluation, and performs initial tests with a Pepper robot.

Findings

Proposed algorithms for social movability evaluation.

Developed a simulator with semantic mapping for social obstacle handling.

Conducted preliminary pushing tests with a Pepper robot.

Abstract

In this paper, we present an in-depth analysis of Navigation Among Movable Obstacles (NAMO) literature, notably highlighting that social acceptability remains an unadressed problem in this robotics navigation domain. The objectives of a Socially-Aware NAMO are defined and a first set of algorithmic propositions is built upon existing work. We developed a simulator allowing to test our propositions of social movability evaluation for obstacle selection, and social placement of objects with a semantic map layer. Preliminary pushing tests are done with a Pepper robot, the standard platform for the Robocup@home Social Standard Platform League, in the context of our participation (LyonTech Team).