Visual Servoing on Wheels: Robust Robot Orientation Estimation in Remote Viewpoint Control

TL;DR

This paper introduces a fast, robust visual servoing pipeline for robot orientation estimation that requires minimal manual labeling and adapts quickly to new environments, enabling rapid deployment of autonomous robots.

Contribution

A weakly supervised pipeline for robot orientation estimation that reduces data labeling effort and accelerates deployment in new environments.

Findings

High tracking accuracy in indoor environments

Deployment in less than 30 minutes in new settings

Effective integration with autonomous navigation systems

Abstract

This work proposes a fast deployment pipeline for visually-servoed robots which does not assume anything about either the robot - e.g. sizes, colour or the presence of markers - or the deployment environment. In this, accurate estimation of robot orientation is crucial for successful navigation in complex environments; manual labelling of angular values is, though, time-consuming and possibly hard to perform. For this reason, we propose a weakly supervised pipeline that can produce a vast amount of data in a small amount of time. We evaluate our approach on a dataset of remote camera images captured in various indoor environments demonstrating high tracking performances when integrated into a fully-autonomous pipeline with a simple controller. With this, we then analyse the data requirement of our approach, showing how it is possible to deploy a new robot in a new environment in less than 30.00 min.