AirTrack: Onboard Deep Learning Framework for Long-Range Aircraft Detection and Tracking

TL;DR

AirTrack is a real-time, vision-only deep learning framework designed for long-range aircraft detection and tracking in UAS, outperforming existing methods and meeting safety standards through innovative image alignment and cascaded classification.

Contribution

The paper introduces AirTrack, a novel deep learning-based detection and tracking system optimized for SWaP constraints and long-range aircraft detection in UAS.

Findings

Outperforms state-of-the-art baselines on AOT dataset

Achieves >95% tracking probability up to 700m range

Successfully passes ASTM DAA safety standards in flight tests

Abstract

Detect-and-Avoid (DAA) capabilities are critical for safe operations of unmanned aircraft systems (UAS). This paper introduces, AirTrack, a real-time vision-only detect and tracking framework that respects the size, weight, and power (SWaP) constraints of sUAS systems. Given the low Signal-to-Noise ratios (SNR) of far away aircraft, we propose using full resolution images in a deep learning framework that aligns successive images to remove ego-motion. The aligned images are then used downstream in cascaded primary and secondary classifiers to improve detection and tracking performance on multiple metrics. We show that AirTrack outperforms state-of-the art baselines on the Amazon Airborne Object Tracking (AOT) Dataset. Multiple real world flight tests with a Cessna 182 interacting with general aviation traffic and additional near-collision flight tests with a Bell helicopter flying towards a UAS in a controlled setting showcase that the proposed approach satisfies the newly introduced ASTM F3442/F3442M standard for DAA. Empirical evaluations show that our system has a probability of track of more than 95% up to a range of 700m. Video available at https://youtu.be/H3lL_Wjxjpw .