Just as the SAE Levels clarified automated driving, so too could a clear taxonomy facilitate software-defined vehicle development. By Megan Lampinen
Everyone is talking about software-defined vehicles (SDVs), but they are not necessarily speaking the same language. SDV has become a buzzword with multiple interpretations, and one company’s understanding of the associated functions and technology can differ widely from another’s. This lack of consensus makes it difficult for automakers and their suppliers to accurately convey features on offer or targets for the future.
“Capabilities and ambitions among OEMs differ drastically when it comes to SDVs,” says Moritz Neukirchner, Senior Director of Strategic Product Management for SDV at Elektrobit. “Some automakers want to build an open app ecosystem and replicate the success of a smartphone’s app store but with vehicle functions. Others want to upgrade the vehicle but don’t care about controlling an ecosystem. Putting it all in one pot really complicates discussions.”
This is where a common terminology or taxonomy could make a huge difference.
What do we call it?
SBD Automotive, a CASE (connected, autonomous, shared and electronic) technology and security consultant to the automotive industry, came up with an early form of internal terminology several years ago, referring to the spectrum of SDVs as ranging from Vehicle 1.0 to Vehicle 4.0. Under this approach, Vehicle 1.0 signals a basic functional vehicle without any form of connectivity. Vehicle 2.0 introduces digitisation, such as infotainment systems and digital cockpits. Updates are possible with Vehicle 3.0, but a vehicle only truly becomes software-defined at Vehicle 4.0, at which point the software running on the car is predominately controlled by a central runtime environment in the cloud.
While this sort of parlance has been useful for SBD, it’s not carried over into the wider industry, and that’s what Elektrobit’s Neukirchner hopes to do with his own version. Shared informally through a LinkedIn post in August 2024 under the description, a “very personal table of ‘Moritz SDV levels’”, it represents a template for a common terminology around SDVs.
Like the SAE Levels for autonomous driving, these also range from 0 to 5. On the Moritz scale, Level 0 is described as ‘software enabled’ and refers to something like parking assist or adaptive cruise control. Level 1 is ‘connected’ and would include mobile phone companion apps and live traffic updates. With Level 2 the vehicle’s features become ‘updateable’ over-the-air, which are still limited to fixes as opposed to introducing new functionality. Level 3 is described as ‘upgradeable’ and opens the door to delivering new features to vehicles already on the road.
“SDV starts with Level 3,” Neukirchner tells Automotive World. “Everything that comes before is essentially maintenance. Today, most automakers are trying to move from a Level 2 to a Level 3, with just a handful of electric vehicle start-ups already on a Level 4.”
With SDV Level 4, the vehicle becomes a ‘software platform’, and the lifecycles of hardware and software are separated. At this level new functions—decided by the automaker—are rolled out to all vehicles in an OEM’s fleet simultaneously, such as the Tesla Boombox. At Level 5 the vehicle becomes an ‘innovation platform’, where third parties provide features and the vehicle driver decides on which ones to install.
For all these levels, Neukirchner also highlights the typical technology enablers. For instance, Level 1 needs a central connectivity unit and Level 3 generally requires zonal architecture. He also flags the business enablers, such as tailored monetisation schemes. “There is a fundamental shift in business mode with the move from Level 4 to Level 5,” he explains. “In Level 4, the OEM is the only business stakeholder: it decides what functions are worthwhile for drivers and serves as the vendor. Level 5 introduces a multi-sided business model where the automaker uses the vehicle as a marketplace to sell functions that a third party develops for end customers.”
An earlier template
2014 saw the publication of SAE J3016 “Recommended Practice: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles”, commonly known as the SAE Levels of Driving Automation. Refined in 2021, it has become the industry standard for referencing driving automation. The vision set out by Neukirchner is far from finalised but represents a similar starting point for the sort of consensus needed to sustain SDV momentum across the industry.
“What we have now with SDV is similar to what we saw early on with autonomous driving,” he says. “Just as not every OEM is trying to reach Level 5 autonomous driving, not every OEM is trying to reach a fully software-defined vehicle. Something like this really helps structure discussions and helps developers articulate where they are today and where they want to go.”
It will also help set realistic industry expectations, and Neukirchner points specifically to the five stages of the Gartner Hype Cycle. This tool was developed to provide reference points for technology evolution and reporting, ranging from the initial ‘technology trigger’ stage through to ‘inflated expectations’, ‘trough of disillusionment’, ‘slope of enlightenment’, and finally ‘plateau of productivity’. “Autonomy has been over-hyped, and expectations were very heightened,” he notes. “We are now on the downhill path in the SDV hype cycle. There are headlines about cutting back on costs and ambitions, but this terminology helps to clarify. Everyone is starting to realise that this journey will not be revolutionary; it’s going to be incremental, and we need to articulate where we stand.”
