Artificial intelligence (AI) algorithms and supporting hardware will be critical in heralding the next stages of automated and ultimately autonomous driving, and a collaboration between Infineon and ZF as part of the EEmotion project demonstrates the viability of this ambitious technology undertaking.
EEmotion successfully integrated AI into the safety-critical functions of the vehicle control system. Source: Infineon Technologies
The EEmotion project aimed to develop an AI algorithm-based control system for automated driving that ensures more precise trajectory control in various driving situations. The project ran from September 2021 to August 2024, was co-funded by the German Federal Ministry for Economic Affairs and Climate Action and had Infineon Technologies AG as the consortium coordinator.
It began by defining the requirements for AI-based functions while aiming to develop AI in control architectures for safety-critical applications. The project also worked on aspects like the development of secure AI-monitored communication, investigation of the simulative development, and taking validation of vehicle dynamics systems into account.
As part of this project, Infineon joined hands with ZF Group to create and implement AI algorithms to develop vehicle control software. These AI algorithms—proven in a test vehicle—controlled and optimized all actuators during automated driving according to the specified driving trajectory.
ZF added AI algorithms to its two existing software solutions cubiX and Eco Control 4 ACC. The cubiX software makes it possible to control all chassis components in passenger cars and commercial vehicles. Next, Eco Control 4 ACC, a predictive cruise control system, was upended using a computationally intensive optimization algorithm and model-predictive control to achieve as much as 8% more range under real driving conditions.
These software solutions with added AI content were implemented on Infineon’s AURIX TC4x microcontroller with integrated parallel processing unit (PPU). This MCU, offering ample computing power, is capable of supporting AI modelling, virtualization, functional safety, cybersecurity and networking functions.
The outcome of loading ZF’s software solutions with added AI algorithms on AURIX TC4x MCU was a demonstration of automated lane changes much more accurately and an energy efficiency boost in adaptive cruise control. This shows how such improvements in driving performance while using lower compute power devices like MCUs could pave the way for cost-efficient Level 2+ assistance systems.
Related Content
- Redefining Mobility with Software-Defined Vehicles
- Arm Highlights Future of the Software-Defined Vehicle
- AI’s Impact on the Current and Future Automotive Industry
- Unveiling the Transformation of Software-Defined Vehicles
- How Automated Driving Is Transforming the Sensor and Computing Market
The post AI algorithms on MCU demo progress in automated driving appeared first on EDN.