Many key technologies, such as autonomous driving, are on the verge of a breakthrough. Practical problems still stand in the way. With the "Learning Chips Lab" research project, three professors at Fachhochschule Dortmund have found a way to solve all these problems. They are currently presenting this approach at the Hannover Messe.
If an autonomous car encounters a motorcycle, the car has to do several things:
- Recognize that it is a motorcycle.
- Know what that means in traffic.
- Assess whether a reaction is necessary.
- If so, decide what the best reaction is,
- and react.
And all this constantly, within fractions of a second and even if there are other cars, cyclists and people on foot nearby. This is possible thanks to the latest developments in artificial intelligence.
One of the most important aspects of this is absolute reliability. So if the internet connection were to weaken from time to time and the calculations were to get stuck in the cloud while the car and all the other vehicles continue to speed through the area, that would be a bad thing.
Independence from the cloud
The current dependence on the cloud is a key weakness in autonomous driving as well as in many areas of Industry 4.0, energy technology, biomedicine and mobility as a whole. In addition to the unreliable connection, data protection and the enormous hunger for performance are also major problems with cloud computing.
This is where the idea of the Learning Chips Lab comes in. Using the possibilities of machine learning, information technician Prof. Dr. Hendrik Wöhrle is developing processes that enable computers to plough through huge data sets independently, for example to understand what a motorcycle is and reliably identify it in any traffic situation.
These chips are absolute specialists
Together with electrical engineer Prof. Dr. Michael Karagounis, who also designs chips for CERN's particle accelerator, among other things, he is designing computer chips that are optimized for such applications - and that are so powerful and energy-efficient that they do not rely on cloud connections when in use, which means they simply avoid the problems mentioned above.
Finally, computer scientist Prof. Dr. Carsten Wolff from the Institute for the Digital Transformation of Application and Living Domains (IDiAL) adds a special twist to these developments: the new chips must be programmable in such a way that users do not have to familiarize themselves with the complex technology of the chips.
And it's all open source
Contrary to the industry's habit of keeping all details of their achievements secret in order to avoid imitations, the researchers at Fachhochschule Dortmund have opted for the greatest possible openness and are publishing all the results of the research project as open source, supervised by Prof. Wolff. This allows researchers from all over the world to pick up on the results and develop them further, thereby greatly stimulating and accelerating this research.
The "Learning Chips Lab" began in fall 2021 and is funded as a research focus by the Ministry of Culture and Science NRW.