12 Platform one Researchers in Japan have developed a noise-suppression system using AI for more effective victim detection by UAVs during natural disasters (writes Nick Flaherty). The ‘pseudo-UAV’ system cancels the noise of the motors and propellers to search for people more effectively in the aftermath of natural disasters, such as earthquakes. This can lead to early detection of victims, enabling a rapid response. Existing research has mostly focused on UAVs equipped with cameras that depend on images to search for victims and assess the situation. However, relying only on visual information can be insufficient, especially when victims are trapped under the rubble or in areas that fall in the blind spots of cameras. Some studies have focused on using sound to detect trapped individuals, different flight movements, complicates the development of a signal-processing filter capable of effectively removing UAV sound from the mixture,” said Prof Premachandra. Rather than machine learning, the system uses Generative Adversarial Networks (GANs), which accurately grasp different types of data. It was used to understand the various types of UAV propeller sound data. This learned model is then used to generate a similar sound to that of the propellers, called pseudoUAV sound. This pseudo-UAV sound is subtracted from the actual sound captured by the onboard microphones in the vehicle, allowing the operator to clearly hear and thus recognise human sounds. This technique has several advantages over traditional noise-suppression systems, including the ability to effectively suppress UAV noise within a narrow frequency range with good accuracy. Importantly, it can adapt to the fluctuating noise of a vehicle in real time. These benefits can significantly enhance the utility of UAVs in search and rescue missions. The researchers tested the system on a real UAV with a mixture of UAV and human sounds. Testing revealed that while this system could effectively eliminate UAV noise and amplify human sounds, there was still some remaining noise in the resulting audio. Fortunately, the current performance is adequate for a proposal of this system for human detection at actual disaster sites. “This approach not only promises to improve post-disaster human-detection strategies, but also enhances our ability to amplify necessary sound components when mixed with unnecessary ones,” said Prof Premachandra. Artificial intelligence Noise suppression to improve search and rescue but this is hampered by the noise of the propellers. Multiple microphones and noise-cancellation algorithms have been used to isolate the source of victims’ sounds and then apply speech recognition. However, it can be difficult for the UAV remote operator to accurately recognise the victim’s voice from the processed sound. This software also uses predetermined words, while the sounds made by victims may vary, based on the situation. Professor Chinthaka Premachandra and Yugo Kinasada from the Department of Electronic Engineering at the School of Engineering in Shibaura Institute of Technology in Japan used AI to develop a different noise-suppression system. “Suppressing the UAV propeller noise from the sound mixture while enhancing the audibility of human voices presents a formidable research problem. The variable intensity of UAV noise, fluctuating unpredictably with April/May 2024 | Uncrewed Systems Technology The ‘pseudo-UAV’ system cancels the noise of motors and propellers to help find victims faster
RkJQdWJsaXNoZXIy MjI2Mzk4