ROHM develops ultra-low-power on-device learning Edge AI chip

2 mins read

ROHM has developed an on-device learning AI chip (SoC with on-device learning AI accelerator) for edge computer endpoints in the IoT field.

The chip utilises AI to predict failures (predictive failure detection) in electronic devices equipped with motors and sensors in real-time.

AI chips generally perform learning and inferences to achieve artificial intelligence functions and that requires a large amount of data to be captured, compiled into a database, and updated as needed. Consequently, the AI chip requires substantial computing power that consumes a large amount of power. That’s made it challenging to develop AI chips that can learn in the field while consuming low power for edge computers and endpoints to build an efficient IoT ecosystem.

Based on an ‘on-device learning algorithm’ developed at Keio University, this new AI chip mainly consists of an AI accelerator (AI-dedicated hardware circuit) and ROHM’s high-efficiency 8-bit CPU ‘tinyMicon MatisseCORE’.

Combining the 20,000-gate ultra-compact AI accelerator with a high-performance CPU has provided learning and inference capabilities but with an ultra-low power consumption of just a few tens of mW (1000× smaller than conventional AI chips capable of learning). This allows real-time failure prediction in a wide range of applications, since ‘anomaly detection results (anomaly score)’ can be output numerically for unknown input data at the site where equipment is installed without involving a cloud server.

ROHM has said that it plans to incorporate the AI accelerator used in this AI chip into various IC products for motors and sensors. Commercialisation is scheduled to start in 2023, with mass production planned in 2024.

Professor Hiroki Matsutani, Dept. of Information and Computer Science, Keio University, who led the team behind the ‘on-device learning algorithm, said, “As IoT technologies such as 5G communication and digital twins advance, cloud computing will be required to evolve, but processing all the data on cloud servers is not always the best solution in terms of load, cost, and power consumption. With the ‘on-device learning’ research and the ‘on-device learning algorithms’ we developed we aim to achieve more efficient data processing on the edge side to build a better IoT ecosystem.

“Through this collaboration, ROHM has shown us the path to commercialisation in a cost-effective manner by further advancing on-device learning circuit technology. I expect the prototype AI chip to be incorporated into ROHM's IC products in the near future.”

For evaluating the AI chip, ROHM is offering an evaluation board equipped with Arduino-compatible terminals that can be fitted with an expansion sensor board for connecting to an MCU (Arduino). Wireless communication modules (Wi-Fi and Bluetooth) along with 64kbit EEPROM memory are mounted on the board, and by connecting units such as sensors and attaching them to the target equipment it will be possible to verify the effects of the AI chip from a display.