Infineon and ZF collaborate to optimize autonomous driving software and control units through AI algorithms

     In order to realize the automatic following of trucks on the highway, formation driving and automatic lane change of cars in the case of unmanned driving, it is necessary to calculate and execute vehicle motion accurately and quickly. Software and AI algorithms safely control drive, braking, front and rear wheel steering, and shock absorption systems. The more efficient an AI algorithm is, the better it can use the available computing power.

     As a participant in the EEmotion project, Infineon Technologies AG and ZF Group jointly developed and implemented AI algorithms for developing and controlling automotive software. The project is co-funded by the German Federal Ministry of Economic Affairs and Climate Action. The AI algorithms developed in this project have been validated in test vehicles to control and optimize various actuators according to specified driving trajectories during autonomous driving.

     Zf has added AI algorithms to its two existing software solutions, cubiX and Eco Control 4 ACC. These algorithms have been implemented on Infineon's AURIXTM TC4x microcontrollers (MCUS) with integrated parallel processing units (PPUs), resulting in higher AI algorithm efficiency and computational power utilization, which in turn improves driving performance and driving safety. Both companies have demonstrated that their solutions can more accurately implement functions such as automatic lane changes than traditional methods without AI. The energy efficiency of driver assistance systems such as adaptive cruise control has also been improved. The combination of improved driving performance and reduced computing power requirements lays the foundation for cost-effective L2+ driver assistance systems.

     Torsten Gollewski, Head of Research and Development at ZF, said: "The EEmotion project shows that our AI-based algorithms offer our customers new advantages. AI makes it possible to equip products with new features and increase the speed and efficiency of product development."

     Peter Schiefer, President of Automotive Electronics at Infineon Technologies, said: "With the world's leading solutions for semiconductor products, software and services, Infineon helps customers develop their own AI applications. Our AURIXTM TC4x is ideally suited for in-vehicle AI applications, and its parallel processing unit enables the fast parallel data processing necessary for AI, opening a new phase of autonomous and even driverless driving."

     Ernst Stoeckl-Pukall, Head of the Digitization and Industry 4.0 Division at the German Federal Ministry of Economic Affairs and Climate Action, said: "The EEmotion project successfully integrated AI into safety-critical functions of vehicle control systems, which has been validated on the software side, making further progress in advanced autonomous driving possible. This project provides an important impetus to strengthen the innovation and competitiveness of the German automotive industry."