Research Analyzer
← Back ICRA 2026

Acoustic Drone Package Delivery Detection

François Marcoux, Francois Grondin

PDF

AI summary

Key figure (auto-extracted from paper)
The first acoustic algorithm successfully identifies drone package delivery events by tracking propeller speed changes, achieving 96% detection accuracy up to 100 meters.
acoustic sensing drone detection package delivery blade passing frequency neural networks security surveillance

Problem

UAVs are increasingly used to smuggle contraband into restricted areas, yet existing surveillance systems focus on drone presence rather than identifying the precise moment of a package drop.

Approach

A ground-based microphone array feeds audio into a convolutional recurrent neural network that estimates propeller blade passing frequencies, which are then analyzed via a change-point detection algorithm to pinpoint delivery moments.

Key results

  • First acoustic algorithm for drone delivery event detection
  • 97% drone presence detection accuracy
  • 16 Hz mean absolute error for blade passing frequency estimation within 150 m
  • 96% delivery detection rate at 8% false positive rate up to 100 m

Why it matters

Provides security personnel with a low-cost, RF-free method to intercept illicit drone drops in sensitive perimeters like prisons and airports.

Abstract

In recent years, the illicit use of un- manned aerial vehicles (UAVs) for deliveries in re- stricted area such as prisons became a significant security challenge. While numerous studies have fo- cused on UAV detection or localization, little atten- tion has been given to delivery events identification. This study presents the first acoustic package deliv- ery detection algorithm using a ground-based micro- phone array. The proposed method estimates both the drone’s propeller speed and the delivery event using solely acoustic features. A deep neural network detects the presence of a drone and estimates the propeller’s rotation speed or blade passing frequency (BPF) from a mel spectrogram. The algorithm ana- lyzes the BPFs to identify probable delivery moments based on sudden changes before and after a specific time. Results demonstrate a mean absolute error of the blade passing frequency estimator of 16 Hz when the drone is less than 150 meters away from the microphone array. The drone presence detection esti- mator has a accuracy of 97%. The delivery detection algorithm correctly identifies 96% of events with a false positive rate of 8 %. This study shows that deliveries can be identified using acoustic signals up to a range of 100 meters.

Index terms

Surveillance Robotic Systems Aerial Systems: Applications Aerial Systems: Perception and Autonomy

Related papers