Infineon targets Aurix AI vector units for BMS designs

Infineon Technologies is focusing on integrating the vector processing unit into the upcoming generation of Aurix automotive microcontrollers, specifically targeting AI applications for the battery management system. With the company boasting the largest global market share for automotive microcontrollers, the incorporation of the vector unit in the latest TC4x family for AI with ASIL D functional safety represents a significant advancement.

Aurix is built on the 32-bit TriCore architecture featuring its own instruction set architecture. The TC4x family introduces a parallel processing unit equipped with a vector-based single instruction multiple data (SIMD) for AI acceleration, in addition to up to six TriCore v1.8 cores operating in lockstep at speeds of up to 500 MHz.

Christian Feldmann, vice president of technical marketing for automotive microcontrollers at Infineon, highlighted the company's plans for the next generation of Aurix, emphasizing the inclusion of a vector processing unit for parallel processing to seamlessly integrate AI into vehicles, particularly for enhancing the battery management system.

The TC4x family, unveiled in November 2024, is slated for mass production in 2025, catering to clients such as Continental and Marelli. Feldmann underscored the significance of leveraging machine learning and AI for optimizing the vehicle battery management system (BMS).

“A battery pack may comprise up to 10,000 cells, yet manufacturers typically monitor only a fraction of them. To address this limitation, manufacturers incorporate a buffer that consumes energy and diminishes range. With Aurix, we can achieve more precise determination of the charging state, potentially boosting range by up to 20%, thereby reducing the need for larger batteries, ultimately leading to cost savings and greater affordability of electric vehicles,” Feldmann explained.

Furthermore, he emphasized the critical role of AI in assessing the safety status of the battery, enabling more accurate measurement of the state of health (SoH) and potentially extending the battery's lifespan by 30%, while also reducing fast charging durations by 20%. In addition to its application in enhancing battery management, Infineon is harnessing AI to streamline chip layout automation, significantly reducing layout time from 60 days to just 12 days, and employing AI for code development.