STMicroelectronics’ new inertial modules enable AI training inside the sensor
STMicroelectronics has introduced ISM330ISN inertial sensors that contain an intelligent sensor processing unit (ISPU). The company said the power the onlife era: interacting with trained devices, with intelligence moving from “on” the edge to “in” the edge.
The ISM330ISN always-on, six-axis inertial measurement unit (IMU) for movement and position sensing uses its embedded intelligence to deliver unrivalled performance and accuracy for its size and power, claimed ST. Used in IoT and industrial applications, the IMU sets 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), thus saving power at system level. ST’s approach integrates a specialised processor, the ISPU, in a small area directly on the sensor chip, optimised for machine learning applications. This enables the ISM330ISN module to have a 50 per cent smaller footprint and consume 50 per cent less power than a typical co-packaged MCU, claimed the company.
Developers can programme the ISPU using ST’s NanoEdge AI Studio. This tool is used to deploy AI applications on STM32 microcontrollers. The technology is now available for programming the ISPU, enabling users to easily 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, claimed the company, and with no specific data science skills 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, Analogue 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.”
The ISM330ISN has the same 3.0 x 2.5 x 0.83mm package outline as conventional inertial modules. Designers can therefore upgrade products quickly and cost-effectively with no obligation to change an established circuit board layout.
The ISM330ISN is covered by ST’s 10-year longevity programme, which provides long-term availability assurance for product designers and manufacturers.
The ISM330ISN is part of ST’s iNEMO family of IMUs. It contains a three-axis accelerometer and three-axis gyroscope with low-noise sensing performance, and an output data rate (ODR) of 6.6kHz. With the ISPU, the sensor ensures consistently high accuracy while consuming only 0.59mA in combination mode with the accelerometer and gyroscope active.
ST’s ISPU architecture, based on digital signal processing (DSP), is compact and power-efficient, with 40kbytes of RAM, and occupies just 8,000 gates on the sensor die. Performing floating-point operations with single-bit precision, ISPU is suitable for machine learning applications and binary neural networks, claimed the company.
The ISM330ISN is scheduled to enter production in the second half of 2022 and will be available from st.com or distributors. NanoEdge AI Studio enabling the creation of libraries designed for specific ISPU part numbers is available at no charge on ST.com.
ST’s ISPU and the NanoEdge.AI tools have been shortlisted for the Embedded Award at Embedded World 2022, in the Hardware and Software categories respectively (in Nuremberg, Germany, 21-23 June). ST will be exhibiting in Hall 4A-148.
