Contrastive Learning for Enhancing Robust Scene Transfer in Vision-based Agile Flight

TL;DR

This paper introduces an adaptive multi-pair contrastive learning method that improves visual representation for agile quadrotor flight, enabling zero-shot scene transfer and robust real-world deployment.

Contribution

It presents a novel contrastive learning strategy that enhances generalization and transferability of control policies in vision-based mobile robotics.

Findings

Successfully generalizes to unseen environments

Outperforms baseline methods in simulation and real-world tests

Enables zero-shot scene transfer without finetuning

Abstract

Scene transfer for vision-based mobile robotics applications is a highly relevant and challenging problem. The utility of a robot greatly depends on its ability to perform a task in the real world, outside of a well-controlled lab environment. Existing scene transfer end-to-end policy learning approaches often suffer from poor sample efficiency or limited generalization capabilities, making them unsuitable for mobile robotics applications. This work proposes an adaptive multi-pair contrastive learning strategy for visual representation learning that enables zero-shot scene transfer and real-world deployment. Control policies relying on the embedding are able to operate in unseen environments without the need for finetuning in the deployment environment. We demonstrate the performance of our approach on the task of agile, vision-based quadrotor flight. Extensive simulation and real-world experiments demonstrate that our approach successfully generalizes beyond the training domain and outperforms all baselines.