IDE Easily Integrates AI In Arm’d Objects

By Mat Dirjish

The race to inject artificial intelligence (AI) into even the simplest of everyday products is not too easy to start let alone approach a finish line. Traditionally, implementing AI in embedded devices is a long, difficult, and expensive process lasting anywhere from months to years. There’s the tedium of accessing extensive and difficult-to-source data. Addressing a plethora of sensors, interfaces, and the maze of codes and algorithms can take a chunk of time most developers can barely, if at all, afford.

Fortunately, the folks at Cartesiam, a company that creates AI software for embedded systems, see things differently. Today, at the Embedded World conference, Cartesiam unveils NanoEdge AI Studio, which the company describes as “the first integrated development environment (IDE) that enables machine learning and inference directly on Arm Cortex-M microcontrollers (MCUs).”

Essentially, the IDE is an intuitive software tool that allows users of Arm’s low-power, low-cost MCUs to easily and cost-effectively integrate machine learning directly into everyday objects, i.e., industrial machines, IoT and automotive applications, household appliances, etc. Marc Dupaquier, general manager and co-founder of Cartesiam, informs, “NanoEdge AI Studio offers a completely different approach, with a cost- and time-efficient and self-learning AI. It allows any embedded designer to develop application-specific machine learning libraries quickly and run the program inside the MCU right where the signal becomes data. It’s the only solution that can run both machine learning and inference on the MCU.”

NanoEdge AI Studio removes traditional AI barriers. It targets companies that either do not have expert resources in machine learning or want to provide their data scientists with a tool for embedded environments. In practice, NanoEdge AI Studio transforms passive sensors into autonomous agents capable of self-monitoring.

What the IDE Can Do

NanoEdge AI Studio…

  • runs autonomously on the developer’s workstation under Windows or Linux with no data transmissions outside the user’s environment.
  • will automatically test, optimize and calculate the best algorithmic combination among more than 500-million possible combinations, after the developer describes the target environment.
  • provides the choice algorithm as a C library that is easily embeddable in the MCU
  • generates libraries that require only 4K to 16K of RAM, making them the most optimal AI algorithms in the industry.
  • enables the execution of unsupervised learning, inference and prediction on the device edge, opening new classes of small, low-power, low-cost devices to AI.

Step-by-Step

Want to take it for a test drive? Then download a trial version of NanoEdge AI Studio. Read to start working on the next big thing? Full licenses of NanoEdge AI Studio are available for purchase from Cartesiam and from Richardson.

Stay abreast of all the cutting-edge developments in the world of artificial intelligence by signing up for Sensors Daily Newsletters. It’s free, fast, and unbelievably easy. And for a media kit offering unique and effective sponsorship opportunities, contact Michael Mitchell via email.