Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Count Every Rotation and Every Rotation Counts: Exploring Drone Dynamics via Propeller Sensing

Xuecheng Chen, Jingao Xu, Wenhua Ding, Haoyang Wang, Xinyu Luo, Ruiyang Duan, Jialong Chen, Xueqian Wang, Yunhao Liu, Xinlei Chen

TL;DR

This work tackles ground-based, contactless sensing of descending drones by focusing on propeller rotational speed as a unaided yet informative signature. It introduces EventPro, an event-camera-based system with Count Every Rotation for precise, low-latency speed estimation and Every Rotation Counts for inferring internal commands and enhancing external localization through multi-modal fusion. The approach combines distribution-informed preprocessing, geometry-informed motion compensation, and adaptive event representation to deliver $(i)$ sub-4 ms latency and speed estimation error as low as $0.23\%$, $(ii)$ up to $96.5\%$ internal-command prediction accuracy, and $(iii)$ over $22\%$ improvement in tracking when fused with other sensors. Extensive real-world evaluations in drone delivery scenarios demonstrate robust performance across varying speeds, distances, angles, and illumination, highlighting practical impact for precision landing and safe ground operations.

Abstract

As drone-based applications proliferate, paramount contactless sensing of airborne drones from the ground becomes indispensable. This work demonstrates concentrating on propeller rotational speed will substantially improve drone sensing performance and proposes an event-camera-based solution, \sysname. \sysname features two components: \textit{Count Every Rotation} achieves accurate, real-time propeller speed estimation by mitigating ultra-high sensitivity of event cameras to environmental noise. \textit{Every Rotation Counts} leverages these speeds to infer both internal and external drone dynamics. Extensive evaluations in real-world drone delivery scenarios show that \sysname achieves a sensing latency of 3$ms$ and a rotational speed estimation error of merely 0.23\%. Additionally, \sysname infers drone flight commands with 96.5\% precision and improves drone tracking accuracy by over 22\% when combined with other sensing modalities. \textit{ Demo: {\color{blue}https://eventpro25.github.io/EventPro/.} }

Count Every Rotation and Every Rotation Counts: Exploring Drone Dynamics via Propeller Sensing

TL;DR

This work tackles ground-based, contactless sensing of descending drones by focusing on propeller rotational speed as a unaided yet informative signature. It introduces EventPro, an event-camera-based system with Count Every Rotation for precise, low-latency speed estimation and Every Rotation Counts for inferring internal commands and enhancing external localization through multi-modal fusion. The approach combines distribution-informed preprocessing, geometry-informed motion compensation, and adaptive event representation to deliver sub-4 ms latency and speed estimation error as low as , up to internal-command prediction accuracy, and over improvement in tracking when fused with other sensors. Extensive real-world evaluations in drone delivery scenarios demonstrate robust performance across varying speeds, distances, angles, and illumination, highlighting practical impact for precision landing and safe ground operations.

Abstract

As drone-based applications proliferate, paramount contactless sensing of airborne drones from the ground becomes indispensable. This work demonstrates concentrating on propeller rotational speed will substantially improve drone sensing performance and proposes an event-camera-based solution, \sysname. \sysname features two components: \textit{Count Every Rotation} achieves accurate, real-time propeller speed estimation by mitigating ultra-high sensitivity of event cameras to environmental noise. \textit{Every Rotation Counts} leverages these speeds to infer both internal and external drone dynamics. Extensive evaluations in real-world drone delivery scenarios show that \sysname achieves a sensing latency of 3 and a rotational speed estimation error of merely 0.23\%. Additionally, \sysname infers drone flight commands with 96.5\% precision and improves drone tracking accuracy by over 22\% when combined with other sensing modalities. \textit{ Demo: {\color{blue}https://eventpro25.github.io/EventPro/.} }

Paper Structure

This paper contains 31 sections, 7 equations, 19 figures, 1 table.

Table of Contents

  1. Introduction
  2. System Overview
  3. Count Every Rotation
  4. Distribution-informed Hierarchical Event Preprocessing
  5. Temporal Distribution-informed Event Filtering
  6. Spatial Distribution-informed Propeller Segmentation
  7. Geometry-instructed Event Motion Compensation
  8. Motion Compensation Framework
  9. Warping Function
  10. Geometry-instructed Objective Function
  11. Adaptive Event Feature Representation
  12. Consistency-driven Dynamic Batch Size
  13. Structure-aware Event Batch Downsampling
  14. Every Rotation Counts
  15. Drone Internal Command Inference
  16. ...and 16 more sections

Figures (19)

  • Figure 1: EventPro locates a real-world delivery drone and guides it to land on the airport landing spot.
  • Figure 2: Propeller rotational speed bridges internal flight commands and external flight status.
  • Figure 3: System architecture of EventPro.
  • Figure 4: Workflow of Count Every Rotation.
  • Figure 5: (a) Event generation principle. (b) The high event-generating speed of two typical commercial drones (Drone-Im30t and Drone-IImeituan) even hovering captured by a commercial event camera.
  • ...and 14 more figures