Biomimetic Neural Networks Revolutionize Odor Detection: Award-Winning Poster at DOS 2024
The 8th World Congress of the Digital Olfaction Society (DOS 2024), held on December 5-6 in Tokyo, celebrated remarkable advancements in digital olfaction. Among the event’s highlights, Liujing Zhuang from Zhejiang University, China, was honored with the Best Poster Presentation Award for the study titled “Electronic Nose Using Biomimetic Spiking Neural Network for Fast Odor Source Estimation.”
Zhuang’s innovative work introduced a biomimetic spiking neural network (SNN) for multi-sensor data fusion in electronic noses, significantly improving accuracy and speed in odor source localization. Tested on public datasets and a mobile robot platform, the approach achieved a root mean square error (RMSE) of less than 0.1 m, outperforming traditional single-sensor methods. This advancement offers strong potential for real-time applications in environmental monitoring and autonomous navigation.
Dr. Zhuang stated to DOS “We developed bionic computing methodologies characterized by bio-inspired perception, swift processing, and energy-efficient information management capabilities. From a biological standpoint, the human brain exhibits an extraordinary level of complexity, with an extensive network of neurons collaborating to sustain the organism’s normal regulatory functions. Spiking neural networks (SNNs) convey information via temporal signals comprised of binary events, thereby integrating multidimensional data. This characteristic enhances their capacity to process spatiotemporal information. In comparison to conventional artificial neural networks (ANNs), SNNs present advantages such as increased computational speed and reduced power consumption. These attributes render them particularly suitable for applications in big data-driven Internet of Things (IoT) systems and low-power neuromorphic chip computing, further for digital olfaction technology.”
We look forward to further breakthroughs in upcoming editions of the congress.
