Prof. Ir Rosdiadee Nordin Featured in Science Magazine Article on tinyML
Prof. Ir. Rosdiadee Nordin, Deputy Dean (Internationalisation) at the School of Engineering and Technology, was recently featured in an article on tinyML, published in the online version of Science Magazine, the world's leading outlet for cutting-edge research across all scientific disciplines. Tiny machine learning (TinyML), represents the next generation of Internet of Things (IoT) technology, also known as the Artificial Intelligence of Things (AIoT). TinyML integrates machine learning (ML) models running on low-power, tiny devices such as microcontrollers, enabling real-time AI processing at the edge without relying on cloud connectivity.
Excerpts from the article featuring Prof. Rosdiadee:
"In Malaysia, Rosdiadee Nordin, an engineer at Sunway University, is using tinyML devices to monitor rivers for plastic trash that might hinder the growth of delicate young mangroves. He and a team of volunteers gathered 9,000 images of plastic trash to train AI models until they could classify it, distinguishing clusters of plastic bottles from clumps of plastic bags. They plan to make the data publicly accessible to help track the location, quantity, and types of plastic waste. 'This will help not only [those who pick up the trash], but also local council or environmental agencies to further plan their waste collection activities,' Nordin says."
Nordin’s work extends to Tasik Chini Lake in eastern Malaysia, which supplies the local Indigenous community with both drinking water and fish. Nordin deployed solar-powered water quality sensors to detect pollution in the lake and send that information to tinyML devices, where AI models use it to make predictions about water quality. 'If we make the water quality data accessible [to] the community, they will be able to understand whether the water is safe for them to consume,' Nordin says.
The lake lacks the internet and cellular connectivity needed to broadcast data to the researchers. To avoid having to travel to each device and manually download the data, Nordin had to improvise. He turned to LoRa, a long-range wireless transmission protocol that uses relatively little power and bandwidth compared with Wi-Fi and can work over tens of kilometres.
Prof. Rosdiadee’s feature in *Science Magazine* is a testament to Sunway University’s commitment to impactful, real-world research. His work highlights how innovative AI solutions, like TinyML, can bridge technological gaps and provide practical benefits to environmental conservation and local communities. By leveraging cutting-edge technology to address pressing ecological challenges, Sunway University continues to position itself at the forefront of scientific innovation.
To read the full article, click the link below:
