🤖 AI Summary
This work addresses the challenge of real-time, robust relative localization of high-speed flying targets—such as micro-drones—for autonomous aerial robots in dynamic interception and multi-agent cooperative missions. We propose an end-to-end perception–tracking framework based on onboard 3D LiDAR. Our method introduces a lightweight 3D voxel occupancy mapping technique integrated with a clustering-driven multi-object tracker, effectively suppressing false detections, reducing latency, and balancing accuracy and robustness. Experimental results demonstrate that, within a 20-meter operational range, the system achieves near 100% target recall, sub-0.2 m 3D localization accuracy, and an end-to-end latency of 20 ms. The framework is validated in both high-fidelity simulation and real-world flight tests, confirming its effectiveness under dynamic, cluttered environments.
📝 Abstract
A new robust and accurate approach for the detection and localization of flying objects with the purpose of highly dynamic aerial interception and agile multirobot interaction is presented in this article. The approach is proposed for use on board of autonomous aerial vehicles equipped with a 3-D LiDAR sensor. It relies on a novel 3-D occupancy voxel mapping method for the target detection that provides high localization accuracy and robustness with respect to varying environments and appearance changes of the target. In combination with a proposed cluster-based multitarget tracker, sporadic false positives are suppressed, state estimation of the target is provided, and the detection latency is negligible. This makes the system suitable for tasks of agile multirobot interaction, such as autonomous aerial interception or formation control where fast, precise, and robust relative localization of other robots is crucial. We evaluate the viability and performance of the system in simulated and real-world experiments which demonstrate that at a range of <inline-formula><tex-math notation="LaTeX">$ ext{20} , ext{m}$</tex-math></inline-formula>, our system is capable of reliably detecting a microscale UAV with an almost <inline-formula><tex-math notation="LaTeX">$ ext{100} %$</tex-math></inline-formula> recall, <inline-formula><tex-math notation="LaTeX">$ ext{0.2} , ext{m}$</tex-math></inline-formula> accuracy, and <inline-formula><tex-math notation="LaTeX">$ ext{20} , ext{ms}$</tex-math></inline-formula> delay.