One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay

TL;DR

This paper introduces a method enabling a robot to learn navigation from a single environment traversal, using an interactive world model and augmentation, achieving zero-shot transfer without additional training.

Contribution

It presents a novel approach combining a single traversal-based world model, pre-trained visual features, and augmentation for effective zero-shot navigation transfer.

Findings

Successful zero-shot transfer in real-world environments

Robust navigation despite environmental variations

Effective learning from limited interaction

Abstract

Recently, model-free reinforcement learning algorithms have been shown to solve challenging problems by learning from extensive interaction with the environment. A significant issue with transferring this success to the robotics domain is that interaction with the real world is costly, but training on limited experience is prone to overfitting. We present a method for learning to navigate, to a fixed goal and in a known environment, on a mobile robot. The robot leverages an interactive world model built from a single traversal of the environment, a pre-trained visual feature encoder, and stochastic environmental augmentation, to demonstrate successful zero-shot transfer under real-world environmental variations without fine-tuning.