Global semiconductor solutions provider Renesas Electronics (TSE: 6723), one year after officially announcing the acquisition of embedded AI solution provider Reality Analytics, Inc. The latest advances in learning (TinyML) solutions.

On June 9, 2022, Renesas announced the acquisition of Reality AI in an all-cash transaction. Reality AI's wide range of embedded AI and TinyML solutions can be used for advanced non-visual sensing in automotive, industrial and consumer products, perfectly suited for Renesas embedded processing and IoT products. They provide machine learning with advanced mathematical methods for signal processing, enabling fast and efficient machine learning inference, and can be applied to small MCUs and powerful MPUs. Using Reality AI Tools, a software environment that supports the complete product development lifecycle, users can automatically explore sensor data and generate optimized models. Reality AI Tools includes analysis capabilities to find the optimal sensor or combination of sensors, where to place the sensors and automatically generate component specifications, fully explaining the model function in time/frequency domain.

In just one year since the announcement of the acquisition, Renesas has launched a series of solutions based on Reality AI technology. The following products will be exhibited at the Sensors Converge exhibition held at the Santa Clara Convention Center from June 20 to 22 at the Renesas booth number: 945.

Reality AI Tools is now tightly integrated with Renesas computing products, natively supporting all Renesas MCUs and MPUs through a built-in part selection engine. In addition, automatic context switching between Reality AI Tools and e2Studio, Renesas' embedded development environment, is enabled.

RealityCheck™ Motor Toolbox is an advanced machine learning software toolbox that uses electrical information from the motor control process to enable the development of predictive maintenance, anomaly detection, and intelligent control feedback without the need for additional sensors. It reduces downtime by early detection of small fluctuations in system parameters that indicate maintenance problems and anomalies. The software not only works seamlessly with Renesas MCUs, MPUs, and motor control kits, but is also fully integrated with Reality AI Tools to create, validate, and deploy sensor classification or predictive models at scale. This feature is a predictive model building toolchain that developers can easily access through the Reality AI toolchain, out of the box.

The RealityCheck HVAC Solution Suite is a vertically integrated solution suite for the HVAC sector. The solution is a complete framework that includes hardware and firmware reference designs, a set of pre-trained ML models ready for product design, and a well-defined process for model training, customization, and field testing to meet specific product needs. This will significantly increase the efficiency of the HVAC system.

The Automotive SWS Solution Kit uniquely combines hardware and software to bring a higher level of protection to passengers. The kit comes with a MEMS microphone array that can be integrated in the assembly or placed on the roof of the vehicle. Use the car-standard MCU that can flexibly adjust the placement position, and run AI detection and positioning software on affordable hardware. The AI model accurately detects and classifies different threats from a distance of 1.5 kilometers for sirens, over 35 meters for cars, trucks and motorcycles, and 10 meters for bicycles and joggers. Localization capabilities are also provided by AI models that calculate angles of arrival, predict distances and detect when threats are approaching or moving away.

Customers in various industries have adopted Renesas AI solutions for various application fields. For example, ITT Goulds Pumps is using Renesas AI technology for data analysis. Brad DeCook, director of monitoring and control R&D at the company, said: "The unique capabilities of Renesas' AI technology have helped us develop machine diagnostic technology that can effectively identify equipment failures caused by severe vibration and high temperature."

Sailesh Chittipeddi, executive vice president and general manager of the Embedded Processing, Digital Power and Signal Chain Business Unit at Renesas Electronics, said: “As more and more customers implement artificial intelligence at the endpoint, we believe the convergence of AI and IoT is creating an important An inflection point. Incorporating Reality AI’s unique and powerful technology into our product lineup will help our customers achieve faster and more accurate information processing and response using less computing and power consumption.”

Successful Product Mix

Renesas has incorporated its AI technology into a broad "successful portfolio" of automotive, industrial and IoT applications. Two examples of solutions that benefit from Reality AI technology are motor-controlled vibration detection and efficient and cost-optimized smart heat pumps. These "successful product portfolios" highlight the application scenarios where Reality AI technology can intelligently detect motor abnormalities for preventive maintenance, which can save customers a lot of maintenance costs. Renesas' "Successful Portfolio" is based on products that are compatible and work together seamlessly, while employing a proven system architecture, resulting in optimized low-risk designs and faster time-to-market. Renesas has launched more than 400 "successful product combinations", designed to help customers accelerate the design process and speed up product launch. For more information, please visit: https://www.renesas.com/win.