Development environment eases machine learning on to microcontrollers
NanoEdge AI Studio enables machine learning and inference directly on Arm Cortex-M microcontrollers (MCUs).
The IDE optimizes and tests algorithms for the described target environment and can then provide the selected algorithm as a C library for embedding in a microcontroller. These libraries require between 4kbytes and 16kbytes of RAM and can provide unsupervised learning, inference and prediction on edge device edges.
Marc Dupaquier, general manager and co-founder of Cartesiam, said NanoEdge AI Studio provides a rapid deployment and low overhead solution for edge devices that can include inference.
“The unsupervised techniques implemented by Cartesiam are complementary to ST’s offering: They are particularly suited to applications where what our customers need to monitor is not something they can predict or have already observed before,” said Ricardo De Sa Earp, general manager of the microcontrollers division of STMicroelectronics, in a statement issued by Cartesiam.
NanoEdge AI Studio is available for download.
Related links and articles:
ARM boosts machine learning for MCUs
Eta ships AI processor for sensor applications
GrAI Matter, Paris research gives rise to AI processor for the edge