AI Technology to Upgrade Search and Rescue Drones
Unmanned Aerial Vehicles (UAVs) are instrumental in search and rescue operations during natural disasters like earthquakes. However, existing UAVs face limitations in detecting victims buried under debris due to their reliance on visual data.
To address this challenge, researchers have developed an innovative artificial intelligence (AI) system that effectively suppresses the noise created by UAVs while improving the detection of human sounds. By minimizing propeller noise and enhancing the isolation of human voices, the new AI-based noise suppression device aims to enhance the overall capabilities of UAVs in search and rescue missions.
This groundbreaking technology, developed by Professor Chinthaka Premachandra and Mr. Yugo Kinasada from the Department of Electronic Engineering at the School of Engineering in Shibaura Institute of Technology, Japan, utilizes Generative Adversarial Networks (GANs) to train the AI model on UAV propeller sound data. The model can then generate pseudo-UAV sounds, which are removed from the actual sound recorded by UAV microphones, allowing operators to more effectively hear and identify human sounds.
The system’s ability to adapt to the shifting noise of UAVs in real-time sets it apart from conventional noise suppression devices and significantly enhances the UAV’s effectiveness in search and rescue operations. While the system has shown success in eliminating UAV noise and amplifying human sounds in tests, further improvements are being pursued to address any remaining challenges.
The research published in the IEEE Transactions on Services Computing highlights the immense potential of this AI technology in improving post-disaster human detection strategies and enhancing the effectiveness of UAVs in disaster response efforts. The ongoing efforts of the research team aim to refine the system further and contribute to saving more lives in disaster situations.
