ALTEN & STMicroelectronics have joined forces to boost your Smart IoT deployment on a large scale. These embedded systems perform automated missions, give contextualized alerts and provide high added-value services.

NANOEDGE AI STUDIO Development tool to quickly deploy Machine Learning into high-tech devices. Complete solution to integrate Machine Learning capabilities into STM32-based devices in a quick, easy, and cost-effective way. NanoEdge AI Studio is a Machine Learning tool, which allows users to bring true innovation to their design. In just a few steps, developers can automatically generate an optimized Machine Learning library for their Edge AI solutions, based on a minimal amount of data.

KEY FEATURES & BENEFITS

• Train & infer directly on the target

• No need to send data to the cloud

• Create ready-to-use optimized models

• Reduce cost and leadtime: benchmark millions of ML algorithm combinations in minutes

• Generate optimized anomaly detection, classifi cation and extrapolation libraries without any AI expertise

KEY APPLICATIONS

• Manufacturing predictive maintenance

• Home appliances anomaly detection

• Human detection

• Motor control condition monitoring

END-TO-END ALTEN SUPPORT, CUSTOMIZED TO CUSTOMERS CHALLENGES

ALTEN support from exploration, needs defi nition, production launch to customer support.

OUR OFFERS

Level 1: START Prerequisites:

None NanoEdge AI Starter Kit (Software and Hardware)

Support:

• Setup

• Defi nition of business challenge

• Proof of concept (POC) for a use case

Level 2: BOOST Prerequisites:

Concept validated Smart IoT Prototyping Kit

Support:

• Defi nition of specifi cations for Minimum Viable Product (MVP)

• Proof of value (POV) for a prototype to evaluate ROI

Level 3: RUN Prerequisites:

POC or POV validated Solution industrialization

Support:

• Industrialization plan

• Architecture of the solution

• Deployment plan

• Change management Level 3 Maintenance

TO GO FURTHER (STMICROELECTRONICS & ALTEN)

1/ AWARD-WINNING TECHNOLOGY

NanoEdge AI Studio enjoys international recognition (2019 IoT Solutions Awards & 2019 Best Innovative Technology award). More recently, version 3.1 won Software category award from Embedded World 2022. This technology is full of innovations allowing to deploy Machine Learning solutions on STM32 MCU based on Cortex M with the possibility of managing the learning directly into the microcontroller despite the strong Flash and RAM constraints.

2 /FROM SMART SENSOR TO SMART SYSTEM

As part of the predictive maintenance of cutting tools for an aeronautical player, we were able to develop a smart sensor to directly process the physical effects of the machine to predict drifts and thus prescriptive maintenance actions. Use of embedded AI to process high-frequency vibration data correlated with real-time data from the factory has made it possible to increase the lifetime of cutting tools by up to 30% and to avoid production disruptions.