Some live in CERN’s facilities near Geneva, on the border between France and Switzerland. most complex and sophisticated machines ever made by humans. Its particle accelerators, including the LHC, which is the largest on the planet, are complex, but the detectors that collect information from collisions are even more so.
From the outside it may seem that once these machines are ready to be used, all that remains is to sew and sing, but nothing is further from the truth. Indeed, processing the enormous amount of information generated by particle collisions requires a titanic computational effort.
Luckily, scientists at CERN have come up with a strategy that has proven to work like a charm. In fact, if it weren’t for him, they might not have found the Higgs boson. Broadly speaking, what they have done is develop an automatic learning model that is able to analyze data very quickly Detectors give them to identify things that are really valuable.
Its machine learning model is valuable for autonomous driving
A priori it seems that the knowledge gained from collisions between particles occurring in the CERN detectors has nothing to do with autonomous driving, but in fact, it has a lot to do with it. And it is that the scientists of this organization have realized that it is possible to use an automatic learning model similar to the one they use in the field of particle physics to process and use in real time information inferior knowledge that They collect sensors. an autonomous car.
Its purpose is to respond quickly to the changing conditions imposed by the environment through which the autonomous car moves.
According to the CERN researchers, the most important contribution their technology can make in this area is the ability to make artificial vision highly sophisticated. What they’re planning, neither more nor less, is to ensure that autonomous cars are able to analyze the information collected by their sensors much faster and with greater accuracy. respond quickly To account for the changing conditions imposed by the environment through which the vehicle moves.
CERN technicians are so confident of their technology that they are running a research project in collaboration with Gensect that seeks to develop it enough so that it can be integrated into autonomous cars of the future. This last company is dedicated to fine-tuning the technologies required for this type of vehicle. It sounds good on paper, however that’s not the only thing we know about.
And it is that CERN has confirmed that the hardware being used in its research project involving autonomous cars is essentially the same that it has used for many years in its research in particle physics: FPGA chips (Field Programmable Gate Array, We will not go into details so as not to complicate this article too much, but at least we are interested to know what they are. programmable ICs Extremely versatile.
At that time, the researchers at CERN decided to use them because they could execute complex decision-making algorithms in microseconds, so they were able to deliver their results with minimal latency. These features on paper fit like a glove with the processing hardware required by an autonomous car, so it makes sense that they’ve chosen to go down a path they’re well aware of. We will track them very closely.
cover image: gensect
more information: CERN