Microchip and partners integrate machine learning for AI at the edge
Microchip Technology has partnered with software partners Cartesiam, Edge Impulse and Motion Gestures to simplify machine learning (ML) implementation at the edge. Using the company’s Arm Cortex based 32-bit microcontrollers and microprocessors in its MPLAB X integrated development environment (IDE), allows Microchip to support customers through all phases of their AI/ML projects including data gathering, training the models and inference implementation, the company says.
Cartesiam, Edge Impulse and Motion Gestures have integrated their ML software products into the MPLAB X development environment. Microchip reports that its 32-bit microcontrollers are increasingly used in AI at the edge applications and implemented by embedded system developers. The solutions can be tested using the EV18H79A or EV45Y33A ML evaluation kits.
Cartesiam is a software publisher, founded in 2016, specialising in AI development tools for microcontrollers. Its development environment, NanoEdge AI Studio, allows embedded developers who do not have any prior knowledge of AI, to rapidly develop specialised ML libraries for microcontrollers.
Another partner is Edge Impulse, which provides the end-to-end developer platform for embedded machine learning, enabling enterprises in industrial, enterprise and wearable markets. The platform is free for developers, providing dataset collection, DSP and ML algorithms, testing and highly efficient inference code generation across a wide range of sensor, audio and vision applications. It has integrated Microchip MPLAB X and evaluation kit support.
Motion Gestures, founded in 2017, provides powerful embedded AI-based gesture recognition software for different sensors, including touch, motion and vision.
No training data collection or programming is required, unlike conventional solutions, and it uses advanced ML algorithms to reduce gesture software development time and costs by a factor of 10 while gesture recognition accuracy is increased nearly 100 per cent.
The MPLAB X IDE ML implementations will be featured during the Embedded Vision Summit 2020 virtual conference this week. Attendees can see video demonstrations at the company’s virtual exhibit, which will be staffed each day from 10:30 am to 1pm PDT.
