Model-based development

Digital Twin

With the ever-increasing computing power of computers, it is becoming possible to perform detailed simulations with which a digital equivalent of reality can be realized, a "digital twin." The TALCOM project worked on the further development of a digital twin environment of the TU/e TruckLab. This laboratory setup is used for research into automated driving at a distribution center, and is a scaled version (1:13.3) of reality.

The use of the digital twin environment makes it possible to evaluate controller concepts first in the virtual environment before they are deployed on physical vehicles. Steps have also been taken to be able to perform simulations in which physical and virtual vehicles can be combined. Much work has been done to get the complex software infrastructure in order and to make the various components work together and be flexible for future extensions.

The developed digital twin environment will be used in teaching in the Automotive Technology master at TU Eindhoven. A comprehensive manual has been written for this purpose, in which students are familiarized step by step with vehicle controls, the various sensors and development of controllers for automated driving with several vehicles simultaneously. This manual was first used by about 50 students in 2022.

Within TALCOM, a Digital Twin Academy has also become available through Prespective, its description can be found at infrastructure "Prespective Digital Twin Academy." Where, incidentally, additional information about the "TU Eindhoven: test setup Digital Twin Trucklab" is also included. 

Model Based Development

In the industry, the shift from "problem-based development" to "model-based development" is underway.

The former is characterized by the practical approach. When solving a problem or for new developments, it relies on knowledge and experience of the product in question. In particular, the focus is on being able to implement the solution in the shortest possible time. For products or systems that have a lot of knowledge and experience in the company and are also not too complex, this is an effective and fast way of working. However, the requirements for automotive designs are becoming increasingly stringent. For example:

- heavier functional requirements (such as energy consumption, ease of use)

- covering safety risks

- increasing reliability over the entire lifetime of vehicles.

It turns out that the conventional way of designing is struggling to keep up with this development, especially as many innovations are taking place (such as electrification), lacking knowledge in a company, and systems are interacting more with each other (multidisciplinarity), such as increasing electronic functionalities. Large companies have come up with "model-based development" as an answer to this, which involves describing reality in simulation models. This allows a great deal of knowledge of systems to be gained at an early stage and allows a much more focused approach to an initial prototype, saving time and costs. An additional advantage is that the development process is much calmer, because with problem-driven development it is much more difficult to estimate problems in advance. In this working group, teaching materials were developed that teach how to use modeling and simulation effectively in solving problems or working out product development objectives. Here the focus is not only on the technology itself, but also how best to communicate a simulation practitioner with the business organization. In addition, cooperation has been initiated with Hyster Yale, for example, in order to integrate practical examples into this at a later stage.