“Auto mobile” – self-driving, the term was coined around 1860 to distinguish the emerging steam and motorized vehicles from horse-drawn carriages. However, cars have never really driven themselves – the driver has to steer, set the speed, engage the right gear, signal and carry out many “secondary activities”. For some years now, more and more of these secondary tasks have been automated – automatic gearboxes, rain and light sensors make life at the wheel easier and more comfortable.
Safety features such as lane departure warning and brake assistants actively intervene in the driver’s most important tasks – steering and speed control – go one step further. The next step is autonomous driving, eliminating the active driver, making them just one of the passengers. Using the camera, radar and other systems, the car recognizes its surroundings and decides with the help of AI how it should react to other road users and the environment.
This list follows a clear trend: While early “helpers” affected individual car elements, like windscreen wipers or lights, safety features now impact multiple systems – including steering, acceleration and braking – and autonomous driving will eventually control all vehicle functions. The integration goes beyond the vehicle itself. Occupants and vehicles are constantly connected to an omnipresent data network, enabling data exchange between them and their surroundings. You can see this in action in the video:
Here, the car is, in a sense, just one element of a shared dining experience, like the wine and side dishes the two protagonists select on the vehicle display during the drive to the restaurant.
The growing automation and integration must be mirrored in development and design tools. Every change can affect other vehicle functions, requiring careful scrutiny. The countless dependencies between parts can no longer be managed manually—systems engineering, a system of systems, is the solution.
The first step in systems engineering is defining and testing functions and their interactions. This network of dependencies spans the entire product lifecycle, from design and production to the utilization phase. It ensures that components and changes can be validated against the original requirements at any stage.
This creates a virtual twin of the vehicle almost automatically, which of course also includes the software. Software supports and controls more and more functions in the vehicle, bringing more and more intelligence to the entire system—above all to compensate for the driver’s absence, who is taking care of other things, like a meeting or choosing the right wine.
The software-defined vehicle offers the possibility of integrating new functions at any time – and customers will demand this. Buying a car today means purchasing an almost unchangeable product whose shortcomings you will have to live with for years to come. This is a concept of the past. Bugs in the software can be fixed quickly and distributed to all cars via over-the-air updates.
Analogically, new or improved functions can be rolled out. If research discovers a new anti-lock braking strategy, (almost) no real prototypes are necessary; instead, the function is tested on the virtual twin, implemented in software and transferred to customer vehicles in the field – in the shortest possible time. In this way, all owners benefit from innovations and bug fixes.
Ralf Steck is a freelance specialist journalist covering the digital product life cycle—from CAD to automation to service. He is a qualified engineer in mechanical engineering and has been supporting the market for more than 30 years.