Drone-Enabled Non-Invasive Ultrasound Method for Rodent Deterrence

Unmanned aerial vehicles open new possibilities for developing technologies that support more sustainable and efficient agriculture. This papproaches often degrade in challenging acoustic conditions, while more recent deep learning-based methods struggle to generalize to unseen microphone arrays or to fully exploit the relationships between microphones in an array. We propose Graph-RelNet, a graph-based neural network that extends relation networks with residual graph convolutional layers to regress the direction of arrival (DoA) from generalized cross-correlation with phase transform (GCC-PHAT) features. Experiments on the TIMIT speech corpus demonstrate that Graph-RelNet consistently outperforms a regression-based adaptation of GNN-SSL, while also surpassing conventional steered response power (SRP) with delay-and-sum (D&S) beamforming at medium and high signal-to-noise ratios (SNRs), especially in scenarios with few microphones. At 5 microphones and SNR = 30 dB, Graph-RelNet reduces mean absolute DoA error by up to 45%. The model also shows strong generalization to unseen microphone arrays and remains competitive under low-SNR conditions.

M. Ratković, V. Kovačević, M. Marijan, M. Kostadinov, T. Miljković, M. Bjelić, "Drone-Enabled Non-Invasive Ultrasound Method for Rodent Deterrence" Drones.
Download Paper