Recently, the “Empowering Consumer Electronics Innovation with AI” conference was successfully held in Nanshan, Shenzhen, co-hosted by Huaqiang Electronic Network and the CIOE China Optical Expo, with support from Huaqiang Mall, Huaqiang Cloud Platform, and Huaqiang Electronic Industry Research Institute.Renesas Electronics Marketing Manager Renliang Lv was invited to attend the conference and delivered a wonderful speech titled “Introduction to Renesas Embedded AI Solutions”. The presentation mainly focused on topics such as “Reality AI real-time analysis software and solutions, BYOM AI solutions,” and more.
During his speech, Renliang Lv pointed out: “We are in a fascinating and challenging era. Although the concept of the Internet of Things (IoT) has been proposed for a long time, the deep integration of artificial intelligence is now giving rise to a global trend of AIoT.”
Currently, the simultaneous commercial use of IoT, artificial intelligence, and 5G technology is collectively building a complete platform for big data processing, endowing terminal devices with unprecedented intelligent potential. However, in traditional designs, many “smart” implementations rely on cloud-based data management and decision-making due to the limitations of terminal device processing capabilities. This not only brings issues such as latency, privacy, and bandwidth costs but also restricts applications that require high real-time performance. To seize the opportunities brought by the wave of artificial intelligence, the industry urgently needs to promote a “decentralization” transformation—delegating some decision-making power to terminal devices, enabling sensors and actuators to have local real-time decision-making capabilities.
“We are not lacking data; the key is how to transform massive data into valuable business decisions,” emphasized Renliang Lv. This transformation requires a profound reshaping of product design thinking: embedded engineers need to explore new paths to extract value from complex data and convert it into meaningful terminal decisions, thereby driving smarter and more efficient product innovation.
Renesas Electronics provides a one-stop embedded AI solution, covering comprehensive support from cloud to terminal. In his presentation, Renliang Lv focused on introducing the Reality AI real-time analysis solution and the BYOM AI solution.
Reality AI Real-Time Analysis Solution
Reality AI is a cloud-based edge AI platform that relies on machine learning model development, capable of automatically learning and analyzing data features to generate AI models. At the same time, the platform emphasizes explainable AI characteristics, aiming to enable users to deeply understand the decision logic of AI models. Through explainable AI, users can clearly know the basis for AI system decisions, the reasons for not taking specific actions, the conditions for system success, and the scenarios for failure, thus allowing them to judge whether to trust the AI system at critical moments and to identify the root causes of errors when the system fails.
In the development of edge AI, Reality AI implements two algorithm functions: signal processing (preprocessing) and classification (SVM/NN). Finally, the trained model parameters and computational structure are expressed in C language, forming functions that are then deployed by customers to MCU/MPU chips.
Currently, e2 Studio has integrated Reality AI Tools, providing users with a seamless workflow. In addition to Reality AI Tools, Renesas has also developed RealityCheck MOTOR for embedded predictive maintenance and control feedback, helping customers achieve intelligent management of motor systems. At the same time, the company has launched RealityCheck HVAC and Automotive SWS solutions, where RealityCheck HVAC will significantly enhance the efficiency of HVAC systems, and Automotive SWS will provide a higher level of protection for passengers.
BYOM AI Solution
BYOM (Bring Your Own Model) AI solution is aimed at professional developers, supporting them in designing the required AI models from scratch, including open-source models and code. The BYOM solution provides two e-AI development tools, including the e-AI Translator tool for MCU and the DRP-AI Translator tool for MPU.
The e-AI Translator tool supports various AI frameworks, such as TensorFlow and PyTorch, and can convert the target AI model into the format required for MCU/MPU development environments. Meanwhile, the e-AI Checker tool can calculate the resource occupancy of ROM/RAM and inference time based on the Translator conversion results and the selected MCU/MPU, generating C source code for easy portability and maintenance.
DRP-AI is Renesas’ original AI acceleration engine, capable of independently completing the entire AI inference process without CPU involvement. The RZ/V2 series AI MPU from Renesas is equipped with the DRP-AI accelerator, offering a range of options from entry-level to high-performance to meet the needs of different application scenarios.
Renesas provides one-stop development support for AI applications based on the RZ/V series, including user-friendly AI applications and development guides, significantly reducing customer development cycles.
In the future, Renesas Electronics hopes to explore the infinite possibilities of AI in edge devices together with partners, making intelligence truly accessible.
END
Renesas Electronics (TSE: 6723)
Technology makes life easier, dedicated to creating a safer, smarter, and sustainable future. As a global microcontroller supplier, Renesas Electronics integrates expertise in embedded processing, analog, power, and connectivity to provide complete semiconductor solutions. Successful product portfolios accelerate the market launch of automotive, industrial, infrastructure, and IoT applications, empowering billions of connected smart devices to improve people’s work and lifestyles. For more information, please visit renesas.com
