Research Challenges for a Future-Proof E/E Architecture - A Project Statement

Kugele, Stefan; Cebotari, Vadim; Gleirscher, Mario; Hashemi, Morteza; Segler, Christoph; Shafaei, Sina; Vögel, Hans-Jörg; Bauer, Fridolin; Knoll, Alois; Marmsoler, Diego; Michel, Hans-Ulrich · 2017 · OpenAlex-citations

DOI: 10.18420/in2017_146

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This project statement addresses the escalating complexity and functional power of automotive electric/electronic (E/E) architectures, driven by the trends of autonomy, artificial intelligence, electrification, and connectivity. The authors identify that traditional ECU-centric, signal-based development approaches have reached their limits in managing this complexity, particularly as vehicles transition toward SAE Level 5 automation where fail-operational behavior replaces fail-silent designs. The central research question is how to design a future-proof automotive E/E architecture that ensures safety, supports automation, and integrates intelligence. To address these challenges, the authors propose a four-pillar approach integrating safety assurance, communication infrastructure, service-oriented architecture (SOA), and artificial intelligence. For safety, the project aims to define safety invariants and reliability constraints for vehicle control systems, ensuring fail-operational behavior through rigorous risk analysis and mitigation. For communication, the study investigates Time-Sensitive Networking (TSN) standards to provide a heterogeneous, high-bandwidth, and deterministic Ethernet-based infrastructure capable of supporting mixed-criticality and hard real-time requirements. The SOA component seeks to shift development from ECU-centric to service-centric models, facilitating hardware-agnostic service specification, discovery, and orchestration to reduce complexity and enable reuse. Finally, the AI focus area explores context acquisition and machine learning methods to enable intelligent personal assistants and adaptive comfort functions, while addressing the architectural challenges of data processing and learning placement. The paper outlines specific methodologies and planned evaluations for each focus area. In safety, the authors plan to assess current and future control system architectures and develop a "safety kit" based on best practices. For communication, they propose developing network modeling approaches using object-oriented and logic programming paradigms, synthesizing schedules via Satisfiability Modulo Theories (SMT), and building a prototypical TSN demonstrator with switches and embedded boards. The SOA effort involves creating a service classification scheme, defining behavioral service interfaces, and building an experimental demonstrator for service design and execution. The AI component focuses on evaluating context reasoning and prediction architectures, including deep learning applications for driver behavior recognition and vehicle detection. The significance of this work lies in its holistic integration of these four domains to create a coherent, future-proof architecture. The authors emphasize the interdependencies between the areas: TSN provides the reliable communication foundation for SOA services; SOA offers the abstraction layer for AI functions and safety-critical services; and safety mechanisms must monitor and constrain learned behaviors to ensure system integrity. By providing a catalog of research questions and a structured framework, the paper aims to guide the development of autonomous vehicles that are not only intelligent and automated but also verifiably safe and architecturally scalable.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• generative ai voice assistants