ST’s inertial modules now enable AI training inside the sensor

1 min read

STMicroelectronics has introduced an inertial sensor that contains an intelligent sensor processing unit (ISPU) that’s capable of interacting with trained devices with its intelligence moving to “in” the edge.

The ISM330ISN always-on 6-axis inertial measurement unit (IMU) for movement and position sensing uses its embedded intelligence to deliver improved accuracy. Intended for IoT and industrial applications, the IMU looks to accelerate response time and extend battery life in equipment such as condition monitors for predictive maintenance, as well as battery-operated asset trackers and industrial applications such as robots.

The intelligence built into the ISM330ISN enables smart devices to perform advanced motion-detection functions in the sensor without interaction with the external microcontroller (MCU), saving power at system level.

ST has integrated a specialised processor, the ISPU, in a small area directly on the sensor chip, optimising machine-learning applications. This enables the module to have a 50% smaller footprint and consume 50% less power than a typical co-packaged MCU.

Developers will be able to programme the ISPU using ST’s NanoEdge AI Studio, that many customers have already used to deploy AI applications on STM32 microcontrollers.

This technology is now available for programming the ISPU, enabling users to generate automatically optimised machine-learning libraries. Designing an anomaly detection library with AI learning capability directly inside the ISPU is possible with minimal data and just a few clicks – and no specific data-science skills are needed.

“Intelligence formerly implemented at the network edge, in an application processor, is now moving to the deep edge, inside the sensor,” said Simone Ferri, General Manager of Marketing, Analog MEMS and Sensors Group, STMicroelectronics. “Our ISM330ISN IMU heralds a new category of smart sensors, leveraging embedded AI to handle complex operations such as pattern recognition and anomaly detection with greatly increased efficiency and performance.”

While presenting new AI-powered opportunities, the ISM330ISN has the same package outline as conventional inertial modules. Consequently, designers can upgrade their products quickly and cost-effectively without having to change an established circuit board layout.