HUNT: High-Speed UAV Navigation and Tracking in Unstructured Environments via Instantaneous Relative Frames

2025-09-23

AI Summary

High-speed UAVs face significant challenges in target tracking within unknown, unstructured environments characterized by degraded perception and the absence of global positioning. Method: This paper proposes a perception–control integrated navigation framework based on relative motion, eliminating reliance on global state estimation. Instead, it directly utilizes onboard instantaneous observables—attitude, altitude, velocity, and relative measurements—to define navigation objectives, enabling seamless transition between search and tracking phases. Contribution/Results: For the first time, exhaustive exploration, target acquisition, and dynamic tracking are unified under a single relative-motion model, substantially enhancing system robustness and response speed. Extensive experiments in dense forests, container yards, and real-world search-and-rescue scenarios demonstrate stable autonomous flight and high-precision target tracking under GPS-denied conditions. The framework provides a scalable, real-time solution for complex field operations.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Localization, Mapping, and Navigation
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Tracking, profiling, and countermeasures against them

Abstract

Search and rescue operations require unmanned aerial vehicles to both traverse unknown unstructured environments at high speed and track targets once detected. Achieving both capabilities under degraded sensing and without global localization remains an open challenge. Recent works on relative navigation have shown robust tracking by anchoring planning and control to a visible detected object, but cannot address navigation when no target is in the field of view. We present HUNT (High-speed UAV Navigation and Tracking), a real-time framework that unifies traversal, acquisition, and tracking within a single relative formulation. HUNT defines navigation objectives directly from onboard instantaneous observables such as attitude, altitude, and velocity, enabling reactive high-speed flight during search. Once a target is detected, the same perception-control pipeline transitions seamlessly to tracking. Outdoor experiments in dense forests, container compounds, and search-and-rescue operations with vehicles and mannequins demonstrate robust autonomy where global methods fail.

Problem

• Enables UAV navigation and tracking without global localization
• Unifies traversal and target tracking in unstructured environments
• Operates under degraded sensing conditions during search missions

Innovation

• Unifies traversal acquisition tracking via relative formulation
• Defines navigation objectives from onboard instantaneous observables
• Seamlessly transitions perception-control pipeline to tracking