By combining hardware acceleration, intelligent MEMS IMU sensing, and an easy-to-use development platform for machine learning, Alif, Bosch Sensortec, and Edge Impulse have dramatically simplified the creation of edge sensing applications that classify complex movements and identify gestures. The performance of this solution is up to 40 times faster than similar existing solutions.
PLEASANTON, Calif., November 14, 2022 /PRNewswire/ — Alif Semiconductor, supplier of microcontrollers (MCUs) and fusion processors, today announced a collaboration with Bosch Sensortec and Edge Impulse to combine hardware and software in reference designs that take the complexity out of creating of precision motion sensing products while dramatically increasing performance, reducing power consumption, and establishing airtight security. Applications that benefit include game controllers, predictive maintenance, and condition monitoring of industrial machinery and freight.
Typically, building and training an artificial intelligence/machine learning (AI/ML) framework that consumes dynamic sensor inputs, then deploying the model to a hardware platform and optimizing the model on this specific hardware for speed, accuracy and power efficiency is a very difficult task operation for developers. Alif and its partners have completely removed this complexity by providing reference designs based on Alif’s Ensemble family secure and low-power E3 microcontroller, Bosch Sensortec’s BMI323, an affordable inertial measurement unit (IMU) with precision and energy efficiency, and Edge Impulse. development platform for machine learning.
The E3 microcontroller uses dual processing domains that uniquely accelerate machine learning. The E3’s high-efficiency processing domain is supposed to be always on to sense the environment, while the E3’s high-performance processing domain wakes up as needed to quickly run heavy workloads and goes back to sleep. In the reference design, the BMI323 IMU provides precision acceleration and angular velocity data to the MCU. Edge Impulse’s platform is used to quickly train an ML model to identify complex continuous gestures such as multiple characters, symbols, and anomalies in repeating complex motion patterns.
The results are amazing.
In less than an hour, an ML model can be created, trained, and deployed on the E3 MCU to detect and identify a complex multi-directional motion pattern and translate it into a symbol. At runtime, the time to complete the inference operation needed to identify the gesture pattern is only 280 μsec.
In comparison, a traditional model trained to identify movement in a single direction deployed on a traditional Cortex-M7 MCU required 2300 μsec for the inference operation to complete, and a traditional Cortex-M4F MCU required 12 msec. These results are 9x slower and 43x slower than Alif’s results, and the simplistic nature of detectable motion is very limiting for real-world use cases.
“It’s really an incredible improvement in performance and potential,” said Stefan Finkbeiner, CEO of Bosch Sensortec. “The increased inference speed translates directly to lower power consumption at the system level and could also be used to make this type of solution work at significantly higher symbol rates. This opens the door to the deployment of smart sensors in a way that is not feasible today.”
“Edge Impulse is purpose-built to leverage the power of advanced tools like processors from Alif and sensors from Bosch Sensortec, enabling breakthrough development without the complexity and timescales traditionally required for ML applications,” said Zack Shelby, CEO and co-founder of Edge Impulse. “The performance demonstrated with this combination is going to be very attractive to companies looking to build the next generation of technology solutions.”
“A key driver of the Ensemble family architecture has been to deliver a greater degree of high-level, integrated functionality to edge platforms than ever before,” said Reza Kazerounian, chairman and co-founder of Alif Semiconductor. “Being able to handle this level of performance of machine learning workloads within the power and price budget of an MCU is unmatched in the industry today.”
For more information, visit the Bosch booth (#C3.322) at Electronica on November 15-18, 2022and our website at www.alifsemi.com.
About Alif Semiconductor:
Alif Semiconductor was founded in 2019 with a vision to meet the growing market need for AI-based, scalable and connected, truly energy-efficient embedded computing solutions. This need has led to Alif Semiconductor’s creation of microcontrollers and a new class of embedded controllers, fusion processors, which enable seamless integration of technology into everyday life by unlocking innovative low-power techniques. , unparalleled functional integration, accelerated AI and ML cutting-edge processing, high security, ubiquitous wireless connectivity, and operating system diversity. For more information, visit www.alifsemi.com.
About Bosch Sensortec:
Bosch Sensortec GmbH, a wholly owned subsidiary of Robert Bosch GmbH, develops and markets a broad portfolio of sensors and microelectromechanical systems (MEMS) solutions suitable for smartphones, tablets, wearable and hearing devices, AR/VR devices, drones, robots , smart home and IoT (Internet of Things) applications. The product portfolio includes 3-axis accelerometers, gyroscopes and magnetometers, 6- and 9-axis embedded sensors, smart sensors, barometric pressure sensors, humidity sensors, gas sensors, optical microsystems and comprehensive software. Since its creation in 2005, Bosch Sensortec has established itself as the leader in MEMS technology in the markets it serves. Bosch has been both a pioneer and global market leader in the MEMS sensor segment since 1995 and has to date sold over 15 billion MEMS sensors. For more information, please visit www.bosch-sensortec.com.
About Edge Impuls:
Edge Impulse is the leading machine learning platform, enabling all businesses to create smarter edge products. Their technology enables developers to bring more ML products to market faster and helps enterprise teams rapidly develop industry-specific solutions in weeks instead of years. The Edge Impulse platform offers powerful automation and low-code capabilities to make it easy to build valuable datasets and develop advanced ML with streaming data. With over 40,000 developers and partnerships with top silicon vendors, Edge Impulse provides a seamless onboarding experience to confidently validate and deploy on the largest hardware ecosystem. To learn more, visit edgeimpulse.com.
SOURCE Alif Semiconductor