Software for embedded systems
Develop Edge AI faster
A new integrated development environment will now enable machine learning and inference directly on arm cortex M microcontrollers. It was developed by the French company Cartesiam.
According to Cartesiam, system developers working with arm microcontrollers can use the intuitive NanoEdge AI Studio AI development platform to integrate machine learning quickly, easily and cost-effectively directly into everyday products (IoT, home appliances, industrial machinery, automotive and more). Until now, implementing AI in embedded systems has been a lengthy process requiring the expertise of data scientists, months or years of development time, complex, extensive data sets and often expensive special hardware.
»With NanoEdge AI Studio, any embedded designer can develop application-specific libraries for machine learning in a very short time and execute the program inside the microcontroller exactly where the signal becomes data. NanoEdge AI Studio is the only solution that can perform both machine learning and inference inside the microcontroller,« said Joël Rubino, CEO and co-founder of Cartesiam.
NanoEdge AI Studio aims to remove the traditional AI barriers: The development platform is designed for companies that do not have their own machine learning experts or who want to provide their data scientists with a complementary tool for embedded environments.
As characteristics Cartesiam emphasizes:
- NanoEdge AI Studio runs autonomously on the developer's workstation under Windows or Linux. No data is transferred outside the customer's environment.
- After the developer has described the target environment, NanoEdge AI Studio tests, optimizes and calculates the best algorithmic combination from more than 500 million possible combinations.
- NanoEdge AI Studio provides the selected algorithm as a C library for easy integration into the microcontroller.
- The generated libraries require only 4 to 16 K RAM.
- NanoEdge AI Studio enables unattended learning, inference and prediction to be performed at the edge of the device. This makes AI accessible for completely new classes of small, power-saving and cost-effective devices.
NanoEdge AI Studio is available for download at www.cartesiam.ai. A test version, which enables developers to work with the IDE, is also available there. Full licenses for NanoEdge AI Studio can be purchased from Cartesiam and Richardson.
