과학기술정보통신부 산하 정부출연연구기관인 한국재료연구원(KIMS, 원장 이정환)이 차세대 웨어러블 엣지 뉴로모픽 컴퓨팅 반도체 소자를 세계 최초로 개발하며, 몸에 부착하는 인공지능 센서가 확산 될 날도 빠르게 다가올 것으로 기대된다.

▲Schematic diagram of wearable edge neuromorphic semiconductor device application that can be attached to the human body (left), photo of a high-density neuromorphic synapse device array completed on a highly flexible substrate (right)
Applying lithium ion thin film transparent flexible substrate to impose high integration and high flexibility
The Korea Institute of Materials Science (KIMS, President Lee Jung-hwan), a government-funded research institute under the Ministry of Science and ICT, is expected to rapidly bring about the spread of AI sensors attached to the body, as it has developed the world's first next-generation wearable edge neuromorphic computing semiconductor device.
The Korea Materials Research Institute announced on the 19th that the research team led by Dr. Yonghoon Kim of the Energy and Electronic Materials Laboratory succeeded in implementing the world's first next-generation neuromorphic semiconductor device with high integration and flexibility by applying a lithium-ion thin film to a transparent flexible substrate.
This technology is a technology for developing next-generation artificial intelligence semiconductor devices with high integration and high flexibility by combining a two-dimensional nanomaterial synthesized at low temperatures using plasma chemical vapor deposition by forming an ultra-thin film of lithium ions on a flexible substrate, colorless polyimide.
It was confirmed that it showed a high handwriting pattern recognition rate of 95% without any significant performance degradation even in a flexible test with a curvature radius of 10 to 30 millimeters (mm) and 700 bending and straightening cycles.
Most of the existing research on artificial intelligence semiconductor devices has implemented devices based on rigid silicon substrates, making it difficult to implement wearable, mechanically flexible, and highly reliable synaptic semiconductor devices. There was a dot.
The research team was able to develop a highly integrated, highly flexible artificial intelligence semiconductor device by combining the lithium-ion ultra-thin film process that was conducted in prior research with the technology for synthesizing two-dimensional nano-materials at low temperatures below 300℃ using semiconductor process technology on a highly flexible substrate.
The next-generation, highly integrated, highly flexible neuromorphic semiconductor device technology developed by the research team is a neuromorphic semiconductor device technology that does not require the existing von Neumann-style information processing device, CPU, and information storage device, memory, and can perform information processing and storage simultaneously with low power. It can be used as a core component of wearable low-power intelligent sensors and edge computing in the future.
“If this technology is commercialized, it will be able to solve the problems of data increase and processing delay required for big data processing between wearable IoT devices,” said Yonghoon Kim, a senior researcher at the Materials Research Institute and the head of the research team. “It is also expected to be applicable to various fields, including wearable edge computing, new concept AI haptics, and vision sensors, as well as various low-power wearable AI devices.”
The research results were carried out through the major projects of the Korea Institute of Materials Science and the Materials Innovation Leading Project of the National Research Foundation of Korea with support from the Ministry of Science and ICT.
In addition, the results of this study were published on March 24 in the March issue of 'Small Methods (IF: 15.367/ first authors: student researchers Park Byeong-jin and Hwang Seung-kwon),' a world-renowned academic journal published by Wiley.

▲Principal researcher Yonghoon Kim, Senior Researcher at Materials Research Institute, co-researchers Jeongdae Kwon, Senior Researcher Jongwon Yoon, first authors of the paper, student researchers Byeongjin Park and Seungkwon Hwang (from left in photo)