인피니언 HV GaN
반도체 AI 인더스트리 4.0 SDV 스마트 IoT 컴퓨터 통신 특수 가스 소재 및 장비 e4ds plus

Infineon, Automotive MCUs Also Use Edge AI for Machine Learning

기사입력2024.10.29 17:27


ImageMob Edge AI Solutions for the AURIX™ Automotive MCU Family

It is expected that edge AI solutions will be installed in vehicle MCUs to safely and real-time process large amounts of data for autonomous driving and automated driving in the future.

ImageMob, a subsidiary of Infineon Technologies AG (Korea CEO Seung-Soo Lee), has integrated machine learning capabilities into Infineon’s AURIX™ TC3x and AURIX TC4x automotive ASIL-D MCUs.

Now, developers can create powerful ML models for the edge with ImageMob Studio and deploy them on Infineon’s proven AURIX MCUs.

The process begins with creating a machine learning model in ImageMob Studio. Once the AI model is complete, users can choose to deploy it directly to the MCU within the platform.

You can then seamlessly generate code, easily implement machine learning on MCUs, and create sophisticated ML models with step-by-step guidance.

ImageMob Studio also provides a sample project for siren detection, demonstrating model creation and deployment.

Users can also learn how to create acoustic models with the AURIX MCU and microphone shield using code examples.

ImageMob also developed a new regression model that can be used to calculate battery level, health, and usage time.

The AURIX TC4x MCU family provides a seamless upgrade path from the AURIX TC3x ASIL-D automotive MCU family.

This enhanced performance is based on the next-generation TriCore™ 1.8. Additionally, the AURIX TC4x features a scalable accelerator family that includes a parallel processing unit (PPU) and multiple intelligent accelerators to support cost-effective AI integration.

These advancements enable the AURIX TC4x family to achieve enhanced machine learning performance, enabling developers to deploy multiple models simultaneously or more complex models.

For example, the AURIX TC3x can handle basic siren detection, while the AURIX TC4x can handle both siren detection and voice interaction simultaneously.

“Integrating safe and reliable AI capabilities into our microcontroller portfolio is crucial to advancing autonomous driving applications in the automotive industry,” said Thomas Boehm, Senior Vice President and General Manager Microcontrollers at Infineon. “Now supporting AURIX microcontrollers with ImageMob Studio increases accessibility to developers worldwide, which is a clear demonstration of Infineon’s role as a leader in innovation in the industry,” said Infineon.

ImageMob Studio is available at https://www.imagimob.com/studio. Additional information about the Infineon AURIX family is available at http://www.infineon.com/AURIX.