Terrain-aware Low Altitude Path Planning

TL;DR

This paper introduces a real-time low-altitude path planning method for NOE flight using onboard RGB images, combining behavior cloning and self-supervised learning to improve path quality.

Contribution

It presents a novel training approach that integrates behavior cloning with self-supervised learning for enhanced path planning accuracy.

Findings

24.7% reduction in average path elevation

Effective real-time path planning in simulation

Improved path refinement through self-supervision

Abstract

In this paper, we study the problem of generating low-altitude path plans for nap-of-the-earth (NOE) flight in real time with only RGB images from onboard cameras and the vehicle pose. We propose a novel training method that combines behavior cloning and self-supervised learning, where the self-supervision component allows the learned policy to refine the paths generated by the expert planner. Simulation studies show 24.7% reduction in average path elevation compared to the standard behavior cloning approach.