Related Experiment Video
Updated: Oct 10, 2025

14:44
Harmonic Radar Tags for Insect Tracking: Lightweight, Low-cost, and Accessible
Published on: May 13, 2025
1.3K
A Non-Invasive Millimetre-Wave Radar Sensor for Automated Behavioural Tracking in Precision Farming-Application to
Alexandre Dore1,2, Cristian Pasquaretta2, Dominique Henry1,2
1Laboratory for Analysis and Architecture of Systems, Toulouse University, CNRS, INPT, 31400 Toulouse, France.
Sensors (Basel, Switzerland)
|December 10, 2021
Summary
This study introduces millimetre-wave radars for automated animal tracking without tags. This technology offers robust, high-precision monitoring for applications like precision farming and genetic selection.
Area of Science:
- Applied ethology
- Animal behaviour science
- Precision agriculture technology
Background:
- Automated animal behaviour quantification is crucial for applied ethology and breeding programs.
- Current methods often rely on animal tags and high computational power, limiting scalability.
- Environmental sensitivities of existing systems restrict their use in real-world farming conditions.
Purpose of the Study:
- To develop and validate a novel automated tracking system for monitoring untagged animals.
- To demonstrate the system's robustness, precision, and low-resource requirements compared to existing technologies.
- To explore the potential of radar technology for large-scale behavioural phenotyping in precision farming.
Main Methods:
- Utilized millimetre-wave radars for real-time tracking of freely moving, untagged animals.
- Compared radar tracking performance against conventional video tracking in terms of processing power and environmental independence.
- Validated the system by monitoring the movements of 58 sheep in an indoor behavioural test.
- Adjusted radar operating frequency and power for tracking at larger spatial scales in field conditions.
Main Results:
- The radar system provided robust and high-precision tracking with low processing power requirements.
- Radar tracking demonstrated independence from light variations and a lower misdetection rate than video systems.
- New estimators derived from radar data improved behavioural phenotyping of sheep.
- Successful demonstration of field tracking at larger spatial scales.
Conclusions:
- Millimetre-wave radar tracking offers a promising, scalable solution for automated, high-throughput monitoring of untagged animals.
- The technology overcomes limitations of current methods, enabling wider application in precision farming and genetic selection.
- Radar-based behavioural phenotyping can significantly advance animal welfare and breeding efficiency.
Keywords:
animal farmingcomputational ethologyradar sensorsradar signal processingsignal classificationwavelet analysis
