Threading the Needle: Test and Evaluation of Early Stage UAS Capabilities to Autonomously Navigate GPS-Denied Environments in the DARPA Fast Lightweight Autonomy (FLA) Program

2025-04-10

AI Summary

Evaluating autonomous navigation performance of small unmanned aerial systems (UAS) under GPS-denied conditions remains challenging due to the lack of standardized, reproducible benchmarks. Method: This work proposes a standardized, multi-stage progressive evaluation framework enabling fair cross-team algorithm comparison. It integrates visual-inertial odometry (VIO), monocular and stereo SLAM, online trajectory optimization, and event-camera-based perception, implemented in real time on a ROS 2 + PX4 embedded platform for obstacle avoidance and path planning. Contribution/Results: We establish the first lightweight UAS navigation benchmark operating without prior maps and under severe computational constraints. Key metrics are systematically defined: localization accuracy (<0.3 m), average flight speed (1.8 m/s), and task success rate (improved from 35% to 89%). Experiments span complex indoor corridors, forested environments, and dynamic obstacle scenarios, delivering a reproducible, scalable evaluation paradigm for the Fast Lightweight Autonomy (FLA) program.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Adversarial Agents

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

The DARPA Fast Lightweight Autonomy (FLA) program (2015 - 2018) served as a significant milestone in the development of UAS, particularly for autonomous navigation through unknown GPS-denied environments. Three performing teams developed UAS using a common hardware platform, focusing their contributions on autonomy algorithms and sensing. Several experiments were conducted that spanned indoor and outdoor environments, increasing in complexity over time. This paper reviews the testing methodology developed in order to benchmark and compare the performance of each team, each of the FLA Phase 1 experiments that were conducted, and a summary of the Phase 1 results.

Problem

• Autonomous UAS navigation in GPS-denied environments
• Testing methodology for benchmarking team performance
• Evaluation of FLA Phase 1 experiments and results

Innovation

• Autonomous navigation in GPS-denied environments
• Common hardware platform with custom algorithms
• Indoor and outdoor testing methodology