New Industry Products

MEMS Motion Sensor Module with Machine Learning Increases Battery Life

February 13, 2019 by Scott McMahan

STMicroelectronics reports that it has integrated machine-learning technology into its advanced inertial sensors to improve activity-tracking performance and increase battery life in mobiles and wearables. The LSM6DSOX iNEMO™ sensor employs a machine-learning core to classify motion data based on known patterns. The company points out that relieving this first stage of activity tracking from the main processor saves energy and accelerates motion-based apps such as fitness logging, wellness monitoring, personal navigation, and fall detection.

The LSM6DSOX is a system-in-package inertial measurement unit (IMU) featuring a 3D digital accelerometer and a 3D digital gyroscope boosting performance at 0.55mA in high-performance mode. The device enables always-on low-power features for an optimal motion experience for the consumer.

The LSM6DSOX is easy to integrate with popular mobile platforms including Android and iOS, thereby simplifying its use in smart devices for medical, consumer, and industrial markets. Auxiliary outputs and configuration options also simplify the use of the company's inertial sensors in optical image stabilization (OIS). Functions including free-fall, wakeup, 6D or 4D orientation, click and double-click interrupts can enable applications such as user-interface management and laptop protection, in addition to activity tracking.

"Machine learning is already used for fast and efficient pattern recognition in social media, financial modeling, or autonomous driving," said Andrea Onetti, Analog, MEMS and Sensors Group Vice President, STMicroelectronics. "The LSM6DSOX motion sensor integrates machine-learning capabilities to enhance activity tracking in smartphones and wearables."

Devices equipped with ST's LSM6DSOX can deliver a convenient and responsive "always-on" user experience without compromising battery runtime. The sensor module also features more internal memory than conventional sensors. A state-of-the-art high-speed I3C digital interface allows longer periods between interactions with the main controller and shorter connection times for added energy savings.

The LSM6DSOX contains a 3D MEMS accelerometer and 3D MEMS gyroscope and tracks complex movements using the machine-learning core at low typical current consumption of just 0.55mA to minimize load on the battery.

The machine-learning core works in conjunction with the sensor's integrated finite-state machine logic to handle motion pattern recognition or vibration detection. Customers creating activity-tracking products with the LSM6DSOX can train the core for decision-tree based classification using Weka, an open-source PC-based application, to generate settings and limits from sample data such as acceleration, speed, and magnetic angle that characterize the kinds of movements to be detected.

The LSM6DSOX supports main OS requirements, offering real, virtual and batch sensors with 9Kbytes for dynamic data batching.

ST points out that it leverages the robust and mature manufacturing processes already used to produce micromachined accelerometers and gyroscopes to produce the sensor modules. The company used CMOS technology for fabricating the interfaces. Each interface features a dedicated circuit which is trimmed to better match the characteristics of the sensing element.

The LSM6DSOX has a full-scale acceleration range of ±2g/±4g/±8g/±16g and an angular rate range of ±125dps/±250dps/±500dps/±1000dps/±2000dps.

The LSM6DSOX fully supports EIS and OIS applications as the module includes a dedicated configurable signal processing path for OIS and auxiliary SPI, configurable for both the gyroscope and accelerometer. The LSM6DSOX OIS can be configured from the Auxiliary SPI and primary interface (SPI / I²C & MIPI I3CSM).

High robustness to mechanical shock makes the LSM6DSOX the preferred choice of system designers for the creation and manufacturing of reliable products. The LSM6DSOX is available in a plastic land grid array (LGA) package.

Key Features

  • Power consumption: 0.55mA in combo high-performance mode
  • "Always-on" experience with low power consumption for both accelerometer and gyroscope
  • Smart FIFO up to 9Kbytes
  • Android compliant
  • ±2g/±4g/±8g/±16g full scale
  • ±125dps/±250dps/±500dps/±1000dps/±2000dps full scale
  • Analog supply voltage: 1.71V to 3.6V
  • Independent IO supply (1.62V)
  • Compact footprint: 2.5mm x 3mm x 0.83mm
  • SPI / I²C & MIPI I3CSM serial interface with main processor data synchronization
  • Auxiliary SPI for OIS data output for gyroscope and accelerometer
  • OIS configurable from Aux SPI, primary interface (SPI / I²C & MIPI I3CSM)
  • Advanced pedometer, step detector, and step counter
  • Significant Motion Detection, Tilt detection
  • Standard interrupts: free-fall, wakeup, 6D/4D orientation, click, and double-click
  • Programmable finite state machine: accelerometer, gyroscope, and external sensors
  • Machine Learning Core
  • S4S data synchronization
  • Embedded temperature sensor
  • ECOPACK®, RoHS and "Green" compliant

The LSM6DSOX is in full production and available now, priced from $2.50 for orders of 1000 pieces.