Visual Semantic Navigation with Real Robots

TL;DR

This paper introduces a ROS-based framework for deploying visual semantic navigation models on real robots, highlighting performance differences between simulation and real-world environments and aiming to improve embodied agent capabilities.

Contribution

We present ROS4VSN, a framework enabling easy deployment of VSN models on real robots, and provide an analysis of model performance in real-world settings.

Findings

Noticeable performance gap between simulation and real-world testing

ROS4VSN facilitates deployment of VSN models on ROS-compatible robots

Experimental validation with two robots confirms real-world challenges

Abstract

Visual Semantic Navigation (VSN) is the ability of a robot to learn visual semantic information for navigating in unseen environments. These VSN models are typically tested in those virtual environments where they are trained, mainly using reinforcement learning based approaches. Therefore, we do not yet have an in-depth analysis of how these models would behave in the real world. In this work, we propose a new solution to integrate VSN models into real robots, so that we have true embodied agents. We also release a novel ROS-based framework for VSN, ROS4VSN, so that any VSN-model can be easily deployed in any ROS-compatible robot and tested in a real setting. Our experiments with two different robots, where we have embedded two state-of-the-art VSN agents, confirm that there is a noticeable performance difference of these VSN solutions when tested in real-world and simulation environments. We hope that this research will endeavor to provide a foundation for addressing this consequential issue, with the ultimate aim of advancing the performance and efficiency of embodied agents within authentic real-world scenarios. Code to reproduce all our experiments can be found at https://github.com/gramuah/ros4vsn.