Learning Transferable UAV for Forest Visual Perception

TL;DR

This paper introduces a transfer learning pipeline for UAV forest navigation, utilizing simulated data and domain adaptation techniques to improve real-world collision-free flight accuracy.

Contribution

It presents a novel approach combining simulated forest datasets and multi-kernel MMD-based domain adaptation for UAV visual perception.

Findings

Achieved 84.08% accuracy in real-world UAV forest navigation.

Demonstrated effective transfer from simulation to real environment.

Proposed a new dataset for forest trail perception in Unreal Engine.

Abstract

In this paper, we propose a new pipeline of training a monocular UAV to fly a collision-free trajectory along the dense forest trail. As gathering high-precision images in the real world is expensive and the off-the-shelf dataset has some deficiencies, we collect a new dense forest trail dataset in a variety of simulated environment in Unreal Engine. Then we formulate visual perception of forests as a classification problem. A ResNet-18 model is trained to decide the moving direction frame by frame. To transfer the learned strategy to the real world, we construct a ResNet-18 adaptation model via multi-kernel maximum mean discrepancies to leverage the relevant labelled data and alleviate the discrepancy between simulated and real environment. Simulation and real-world flight with a variety of appearance and environment changes are both tested. The ResNet-18 adaptation and its variant model achieve the best result of 84.08% accuracy in reality.