IoT AI specialist SensiML has partnered Infineon to build a complete AI/Machine Learning (ML) solution for the Infineon PSoC 6 family of microcontrollers (MCUs) and the range of sensors they support.

The collaboration combines SensiML’s Analytics Toolkit AI development software with the Infineon ModusToolbox and ultra-low power dual-core PSoC 6 MCUs. The aim is to provide developers with a faster and easier way to record data from Infineon XENSIV sensors, create sophisticated AI/ML-based models, and implement them on PSoC 6 MCUs. Using this approach, designers with little to no data science expertise can add local intelligence to their IoT designs for smart home, industrial, fitness, and other applications.

The SensiML Analytics Toolkit uses labeled datasets to create efficient AI models rapidly. It also enables developers to implement high performance edge algorithms without the need for hand coding or data science expertise.

The toolkit complements the PSoC 6 family, which is based on an ultra-low power architecture suitable for battery-powered edge IoT applications. The PSoC 6 family features dual-core Arm Cortex-M4 and Cortex-M0+ processors, which enable design partitioning for simultaneous power/performance design optimisation. The dual-core architecture also enables users to run their AI application generated by the SensiML tools on one processor, and their application code on the other processor.

“Infineon has a large and loyal following for their innovative and extensive sensors, MCUs, and associated development tools,” said Chris Rogers, chief executive officer of SensiML.  “SensiML complements that offering with AI/ML tools that make it easy for developers to add sophisticated AI functions in a small memory footprint with ultra-low power consumption. With AutoML support covering a broad array of sensor inputs, SensiML is an ideal solution for IoT edge processing across Infineon’s XENSIV portfolio.”

“SensiML offers a quick and easy path for our PSoC 6 MCU, connectivity and sensor customers to add low-power, local intelligence to their edge IoT applications,” said Steve Tateosian, vice president of IoT Compute and Wireless Product and Business Line at Infineon. “Through our mutual collaboration, we can now offer a complete suite of software and hardware development tools and products, enabling developers and Hackathon participants to create next-generation smart endpoint IoT devices for a wide range of end uses.”

SensiML and Infineon are marking the beginning of the collaboration with a “Build AI for the IoT Design Challenge” on Hackster.io.  The contest will challenge innovators to use the SensiML Data Analytics Toolkit and Infineon ModusToolbox software along with Infineon’s XENSIV sensors, Wi-Fi/Bluetooth connectivity ICs, and PSoC 6 MCUs to capture and label sensor data, train AI/ML models, and deploy the resulting AI/ML solution on a PSoC 6 MCU to create intelligent endpoint applications. The goal of the challenge is to highlight how quick and easy it is for developers to use the complete end-to-end development platform provided by the two companies to implement local, low power, sophisticated AI-based IoT edge devices.