On-chip AI, sensor fusion and Qvar touch sensing in a 6-axis IMU
By embedding AI, with sensor fusion and a Qvar electrostatic sensor, the LSM6DSV16X 6-axis inertial measurement unit (IMU) from STMicroelectronics, gives a boost to high-end applications.
The high-performance, low-power IMU features an on-chip Qvar electrostatic sensor to support the user interface by sensing tap, double tap, triple tap, long press, L/R and R/L swipe.
Also integrated in the IMU is a 3-axis digital accelerometer and a 3-axis digital gyroscope. The LSM6DSV16X features a triple-channel architecture which processes acceleration and angular rate data on three separate channels for the user interface, optical image stabilization (OIS), and electronic image stabilization (EIS). The LSM6DSV16X also adds dedicated configuration, processing, and filtering.
In edge computing, configurable motion tracking is enabled by the programmable finite state machine (FSM) for accelerometer, gyroscope, and external sensor data processing with a 960Hz data rate.
The machine learning core (MLC) enables context awareness with exportable features and filters for AI in IoT applications.
The adaptive self-configuration (ASC) allows the FSM to automatically reconfigure the IMU in real time. This is based on the detection of a specific motion pattern, or on the output of a specific decision tree which is configured in the MLC, and needs no intervention from the host processor.
The analog hub in the LSM6DSV16X IMU can connect to an external analog input and convert this into a digital signal for processing. The standard interrupts offered by the IMU are for free-fall, wake-up, 6D/4D orientation, click and double click.