Bosch has developed the first self-learning sensor for wearables that it says can update its machine learning edge AI framework without access to the cloud or a smartphone. This can help extend battery life.
The BHI260AP self-learning edge AI sensor System in Package (SiP) enables manufacturers of wearable and hearable devices to provide highly personalized fitness tracking through the self-learning AI software in the sensor. It recognizes and adapts to a wide variety of movements and is able to learn any new fitness activity that is based on repetitive, cyclical patterns.
“The self-learning AI sensor will change how users interact with their fitness devices from a mere one-way approach to an interactive way of training,” said Dr. Stefan Finkbeiner, CEO at Bosch Sensortec. “This new sensor combines Bosch Sensortec’s long-term experience in smart motion sensors with its strong competence in innovative software development.”
Bosch has been a key partner of French AI software developer Cartesiam which allows this edge AI capability.
The Bosch self-learning AI software is available with a standard set of more than fifteen pre-learned fitness activities based on classification libraries, so no training is required before use. It also offers four product features: learn, personalize, auto track and enhance.
The learning mode offers users the option to add new fitness activities that were originally not supported, enabling them to customize the device to their individual needs. The personalization feature enables users to adapt existing, pre-learnt activities to their own individual style, increasing the accuracy of calorie counting and activity analyses.
With the auto track function, users can automatically track fitness activities without any manual intervention and analyze their intensity with activity type and count over time – enabling both endurance and strength training. Finally, manufacturers can add new fitness activities without having to modify the software or needing an original dataset. These new transferable exercises may be provided by coaches or star athletes, enabling benchmarking against the best and learning from experts, or simply from the users’ friends. This enhances the perceived value of devices and strongly helps the manufacturers to differentiate.
Next: Edge AI boost for battery life
Avoiding the need for cloud access or even a smartphone keeps the data private and means activities can be continuously tracked and analysed. Edge AI also minimizes latency and power consumption, meaning users can get fast, real-time feedback and extended charging intervals on their devices. The 1.8V SIP uses 249 µA for self learning at 25Hz and 386 µA at 50Hz, leading to a trainng time of 30s. The standby current is 8 µA.
The sensor SIP includes the hardware, software and embedded AI to reduce development time and costs and cut time to market. In addition to the self-learning AI software, Bosch Sensortec offers a wide range of software solutions for the BHI260AP. Wearables manufacturers can simply load the required software on to the sensor to enable solutions targeted at specific use cases including orientation tracking, position tracking (PDR) and swimming.
The sensor recognizes four different swimming styles: butterfly, backstroke, breaststroke and freestyle. Since the BHI260AP is a fully programmable sensor, manufacturers can build their own customized software to be embedded in the BHI260AP or upload customized solutions depending on the users’ context. To protect these customized solutions from unauthorized use, the smart programmable sensor offers a digital signature.
An application board with a BHI260AP shuttle board can be wirelessly connected over Bluetooth Low Energy (BLE) to smartphones for development.
The BHI260AP self-learning AI sensor and the corresponding shuttle board are available via Bosch Sensortec’s distributors. The different software packages are available for download at the Bosch Sensortec website.
Related edge AI articles
- NANOEDGE AI STUDIO OPTIMIZED FOR THE STM32 DEV BOARDS
- ETA COMPUTE PIVOTS WITH SYNAPTICS BACKING
- TOOLS TACKLE BIAS IN EMBEDDED AI SYSTEMS
- BLAIZE LAUNCHES CODE-FREE TOOL FOR EMBEDDED VISION AI
Other articles on eeNews Europe