Bosch Sensortec’s self-learning AI sensor targets activity tracking

Bosch Sensortec has introduced a new sensor targeted at applications that require always-on sensor processing at ultra-low power consumption. A software framework allows the offering of all features as virtual sensors which can be enabled/disabled according to end-user's requirements. 

The BHI260AP smart sensor includes a 32-bit customizable programmable controller, a 6-axis IMU, and a wide variety of software functionalities. The overall size of the system-in-package with an accelerometer, gyroscope and a programmable 32-bit microcontroller, is 3.6 X 4.1 mm.

Bosch Sensortec's self-learning sensor packs an accelerometer, gyroscope and a programmable 32-bit microcontroller, into a compact 3.6 X 4.1-mm footprint.

The most unique aspect of the design is the hardware-software co-design of AI and non-AI components, which allows a single smart sensor to be used to execute a different firmware function depending on the user’s context.

The smart sensor offers many state-of-the-art functions that are useful for wrist and head mounted devices, including dedicated solutions for pedestrian dead reckoning, swimming analytics and fitness tracking, and generic software platforms with self-learning AI software and orientation estimation. The orientation estimation platform can be used for building applications such as position tracking, whereas the self-learning AI platform for building applications involving tracking repeating patterns during usage for a simplified human machine interface or enabling a personalized recommendation system.

The sensor features a set of software algorithms, including:

  • Self-learning AI software for fitness tracking. This software enables on-device learning and automatic tracking of a wide variety of fitness movements, including options for on-device individual-specific personalization of movements and support for increasing numbers of activities without the need to modify the original software. 
  • Swim analytics. This software dedicated for wrist wearables generates useful information about users' swimming activities, such as length count, style of swimming and stroke counts. 
  • Pedestrian dead reckoning. This software helps to reduce the power consumption of a wearable device by enabling the duty cycling of power-consuming GNSS components, as well as improving the accuracy of outdoor positioning with pedestrian dead reckoning.
  • Relative and absolute orientation. This software estimates relative and absolute orientation of the device, including outputs such as rotation vector, game rotation vector, linear acceleration and gravity.

For evaluation, custom design and speeding up time to market, the sensor comes with a software development kit for customizing / programming the sensor and a shuttle board attached to an application board, which can be connected wirelessly to companion apps on smartphones. The sensor also comes with a software development kit for programming and adding custom code making use of these platforms and dedicated solutions. Bosch Sensortec’s online community can also help to answer specific questions from developers

The product is in sampling for early customers.

Editor’s Note: Bosch Sensortec will be speaking about its new, self-learning AI sensor at Sensors Innovation Week, a digital event series taking place November 16-18, 2020. For more information and to register for your free pass, click here.