The Software Defined Vehicle is leading to a fundamental overhaul of in-vehicle E/E architectures – a topic that the VDI Congress ELIV, as the industry’s most important international gathering, has been addressing for years. Keywords such as standardization and open-source software are shaping the discussion. ELIV 2026 will provide the latest insights on this topic with a preview of a new white paper from ZVEI e.V. Co-author Michael Niklas-Höret, SDV Software Architect at AUMOVIO, will present the paper at the congress and answer our questions in advance.
Mr. Niklas-Höret, you will be presenting a highly anticipated ZVEI white paper at ELIV. How did this paper come about, and what questions served as the catalyst for this project?
Michael Niklas-Höret: The initial impetus came directly from our member companies. Every market participant is acutely aware that the supplier business is undergoing a fundamental transformation due to the new E/E architectures. Against this backdrop, we began at the end of 2023 to structure our initial strategic ideas regarding the impact of new E/E architectures on suppliers; the intensive core work then continued through 2025. Numerous industry experts participated in these in-depth, pre-competitive analyses.
In doing so, we deliberately did not limit ourselves to theoretical considerations but instead ran through concrete use cases. For example, we analyzed complex functions such as Level 2+ autonomous driving, modern X-by-wire systems, and critical processes related to key management and vehicle access. We projected these specific functions onto modern server-zone architectures as examples to identify technological pain points and assess their impact on the entire supply chain.
The announcement for your talk mentions a “centralization of logic.” How would you describe the technological core of this transformation?
Michael Niklas-Höret: A fundamental shift in the intelligence architecture is evident. In the past, specific, functional domain knowledge was firmly embedded in decentralized control units. A supplier would design a dedicated hardware box, equip it with its software logic, fully secure the entire system, and deliver a fully functional package to the OEM. This familiar structure is now breaking down. In the future, the logic for high-level behavior will migrate to highly integrated systems based on server-zone architecture, such as zones or high-performance computers (HPC).
As a result, traditional mechatronic edge devices on the vehicle’s periphery are losing their local intelligence and being drastically simplified. In the future, suppliers will need to provide the domain expertise they have built up over decades as decoupled software components for higher compute layers. At the same time, many automakers are attempting to build up this knowledge internally and implement it themselves as part of the shift toward software defined vehicles.
In practice, where do you see the biggest challenges for development engineers?
Michael Niklas-Höret: Functional verification is becoming a truly Herculean task. When a traditional function is logically divided, a complex chain of effects emerges, extending from the input signal to the mechanical response. However, the physical and regulatory requirements for response times, safety, and security remain exactly the same. Proving that a function operates absolutely error-free when it is distributed across multiple control units and central servers requires entirely new validation concepts. For example, the logic governing high-level behavior can only be shifted into highly integrated systems to the extent that the technical solution can continue to meet the requirements of the chains of effects.
Added to this is the issue of power management. Centralized architectures require a rethinking of power distribution, while at the same time the system must respond in a fraction of a second. In this white paper, we analyze in detail the use of eFuses and new power modes, since smart, centrally controlled power distribution can reduce power consumption and significantly improve range – especially in electric vehicles.
A key point in your preliminary publications is the call for standardization and a thoughtful approach to open-source solutions. How can this be reconciled with OEMs’ desire for a distinctive brand identity?
Michael Niklas-Höret: The current market situation calls for radical honesty. My personal view is that automakers set the bar far too high for what actually sets them apart from the competition. In the automotive industry, we are still building too many expensive, isolated solutions for features that remain hidden from the end customer and offer no tangible added value. This isolated development required significant resources, the benefits of which are not immediately apparent to the buyer.
Instead, as suppliers, we envision a consistent harmonization of semantic interfaces and data formats in the strictly non-differentiating area. If ECU manufacturers do not have to set up a separate matrix from scratch for every model and every OEM, development efforts will decrease rapidly. Such standardization, supported by tool-based allocation methods, is absolutely essential to keep the ever-increasing complexity under control and ensure cost efficiency.
Development costs are one factor; speed is another. What are the benefits of standardization in this regard?
Michael Niklas-Höret: At its core, this is about optimized reuse – that is, the smart reuse of components across different vehicle generations and manufacturers. This engineering culture is already evident among Chinese market players: If an existing design meets 80 percent of the requirements of a new use case, it is used, and only the remaining portion is adapted. The reuse of code or hardware is considered a mark of distinction and an honor there. European players, on the other hand, tend to rule out the 80-percent solution, opting instead to spend years perfecting a 100-percent new development. That costs time and resources that we simply no longer have in the face of global competition.
This inevitably brings us to the topic of lifecycle management. Will software play a decisive role in determining how long a vehicle remains attractive to the user in the future?
Michael Niklas-Höret: Absolutely right. The era of static “feature cars,” in which the customer purchased all available features outright at the time of purchase, is over. Highly integrated computer systems are paving the way for new features that work across vehicles or build on existing ones. In the future, vehicles will be continuously updated “over the air.” This has a massive impact on the vehicle’s residual value in the used car market, especially when buyers can add or remove features at will. Many people immediately think of monetized gimmicks like subscribing to heated seats or other vehicle features – but that’s an oversimplification. Furthermore, vehicle connectivity makes it necessary to continuously roll out highly critical security upgrades. In addition, regulatory authorities will in the future require adjustments for fleets in operation that can be implemented exclusively through software updates.
It will be interesting to see how acceptance among buyers develops. Whether European customers are willing to purchase a vehicle with good basic features that only reaches its full potential through later updates remains one of the most intriguing market questions for the coming years.
To wrap up, let’s take a look at the upcoming ELIV: The white paper analyzes the current situation. What key messages can participants expect from your in-session keynote?
Michael Niklas-Höret: The most important message of our pre-competitive work is a clear commitment to cooperation. Suppliers cannot shoulder the enormous burden of functional validation in distributed architectures on their own. The white paper is intended to serve as a solid foundation for entering into a very concrete dialogue with vehicle manufacturers. The goal is to jointly define the ecosystems of tomorrow, address the high cost pressures, and ensure that both sides of the value chain remain profitable.
The white paper includes specific recommendations for action: among other things, the widespread use of tool-based systems for allocating software to server zones, as well as a bold shift away from developing non-differentiating vehicle parts in favor of industry-wide standards. Technological evolution waits for no one. If the ecosystem of suppliers and OEMs is to continue evolving in a way that is fit for the future, it is crucial to shape the transition to the Software Defined Vehicle now – jointly and at a significantly accelerated pace.
