Implementing Edge AI in Software Defined Vehicle Motion Control

The evolution of software-defined vehicles demands complex interaction between hardware, software, and data. Traditional automotive architectures rely on static control units that lack the capacity to process data or execute artificial intelligence algorithms locally at the vehicle edge.

Furthermore, conventional development methods depend on large-scale data logging, which increases system complexity, creates data management bottlenecks, and extends time-to-market for original equipment manufacturers (OEMs). To achieve continuous optimization, automotive systems require a transition toward dynamic architectures capable of local learning and adaptation.

Integrating Edge AI and Cross-Domain Control Units
To resolve these infrastructure bottlenecks, a global partnership has introduced edge AI technology into vehicle motion control solutions. The architectural solution combines centralized and zonal cross-domain control units from Schaeffler with specialized AI-infrastructure software. This design integrates real-time data collection and model deployment software directly into the control hardware.

The technical configuration spans multiple vehicle domains, including the powertrain, energy systems, chassis, and body components. By embedding these capabilities, the system establishes a flexible framework for next-generation centralized vehicle architectures.

"Our control units run data-driven and AI-based functions within the vehicle, enabling the next generation of vehicle architectures," explains Thomas Stierle, CEO E-Mobility at Schaeffler AG.

Technical Features and Lifecycle Management
The integrated software layer features two primary operational components: a targeted real-time data collection tool and an AI model lifecycle management director. The data collection software allows localized, selective data extraction without requiring extensive continuous logging, reducing onboard processing overhead. Simultaneously, the lifecycle management component permits engineers to deploy and update AI models directly on the active vehicle edge.

This configuration transforms static electronic components into adaptive platforms. "Hardware centralization is the first step; the software-defined vehicle is realized when AI can be running at the edge and when hardware continuously learns and adapts," notes Jeff Chou, CEO and Co-Founder of Sonatus.

Integration Benefits for Automotive Manufacturers
The implementation of pre-integrated software infrastructure within central control units simplifies the development process for automakers. This configuration enables the execution and optimization of core motion control functions, such as steering, braking, and energy management, directly on the control hardware.

"Our central control units are equipped with a pre-integrated software infrastructure," states Rodrigo Peres, Senior Vice President of Business Unit Vehicle and Battery Controls at Schaeffler AG. "This significantly simplifies integration for our customers and helps them accelerate the centralization of their software architecture."

By facilitating targeted data collection, the platform provides deeper insights into vehicle performance and faster resolution of operational anomalies. Consequently, automotive manufacturers can manage escalating system complexity, accelerate hardware-software development cycles, and implement performance updates throughout the vehicle lifecycle without modifying physical components.

Edited by Maria Brueva, Induportals editor – adapted by AI.

www.schaeffler.com