Since the name “Zhuoyu” is very difficult to pronounce in Europe, they decided to start here with the three letters ZYT. ZYT belongs to the still-young full-service providers in autonomous driving that are already represented under many brands in China and are now taking the leap to Europe. We were able to experience a first largely autonomous lap from Messe München to Bogenhausen and back again – in a Hongqi EH7. It had previously been trained on the routes for four weeks and drove the route almost flawlessly: Twice it couldn't decide whether to overtake behind a bus and a truck: The truck parked in the second lane identified the Hongqi as “standing at the traffic light” and waited for it to start moving again. At the bus it spontaneously wanted to pass when the bus indicated with its blinker and pulled out of the stop.
In both cases, the systems here still lack the “feeling” that humans have. But what AI has already managed to train: to take an on-ramp in such a way that it does not make all passengers feel sick by not strictly following the boundary lines, but interpreting the radius a bit more freely and then accelerating properly on the on-ramp, which is very different from China, where traffic is usually from jam to jam or glides, even on major highways.
One still has to adjust in the typical situation of “standing cars” at the end of a traffic jam: Here the safety driver braked hard when he perceived that the vehicles ahead of us suddenly slowed down extremely and our Hongqi had not yet reacted. Otherwise the vehicle drove the sometimes complex route, on which there were also nasty road surface irregularities and hardly marked bus lanes, very confidently!
Then we wanted to know more and delved deeper in an interview with ZYT Germany chief
Nils Oldemeyer.
How is autonomous driving developing from your perspective?
Nils Oldemeyer: Very good and dynamic! It all started with robotics, that is, strictly rule-based. But with generative AI there was a major upheaval, and we are now developing data-driven. That means the system is self-learning. You can imagine it like a child that is repeatedly rewarded for its efforts and successes and thus learns. Of course we do not need to reward the AI, but we can always give it new learning loops.
What led to the decision to use a Hongqi vehicle today?
Oldemeyer: That was availability and a bit of luck: The Hongqi EH7 comes with ten cameras, a radar, and the new Qualcomm 8650 chip. The car actually arrived in Duisburg only four weeks ago by train, and in the meantime we uploaded the data and then began to “train” here. It was also interesting for us to see how much the vehicle can teach itself in this time.
Can't the data and “learnings” simply be transferred from China?
Oldemeyer: No, the data may not simply be transferred to other countries or regions. They are stored for Europe in a local data center. And: Germany has completely different driving situations than China, and these must be learned. Therefore, not the amount of kilometers driven and data is decisive, but the quality of the data. The aim is to find as many “pink elephants” here as possible, such as sign bridges or poorly marked road markings – i.e., special and challenging driving situations that must be learned.
And how long does one have to train there?
Oldemeyer: We want to be driving in Europe in the second quarter of 2026 and be able to offer demos.
Does Zhuoyu offer only complete systems or also individual components?
Oldemeyer: That depends on the OEM. What is certainly unique about us:
We offer a complete software stack. Typically the OEM decides which chipset it uses – that can be Qualcomm, Nvidia, or Texas Instruments. We then basically sell the box with the chipset, and the software stack is tailored for the OEMs.
In what sense?
Oldemeyer: Each OEM has particular preferences as to how the cars should behave. Some higher-performance brands or models may perhaps need a somewhat “longer leash.” In general 80 to 90 percent are the same, but there are differences in driving performance. We work this out in collaboration with the OEMs.
How long does it take for such a software stack to be trained?
Oldemeyer: Our “Europe trainings” we want to complete by mid-2026 – so roughly speaking, that takes about a year. This also depends on the conditions and the fleet. We started with four cars, will soon double to eight, and in the end will have 15 vehicles in operation. In principle the cars adhere very well to all rules, but in fine-tuning it is about how you drive as a person: that you perhaps approach a highway on-ramp or off-ramp that closes in on you a bit further out. It is also important to accelerate quickly on German highways and to know that very fast motorcycles can be on them. Or that buses or trucks are reliably detected. They are larger in Europe and the USA than in China, where there is mostly distribution traffic in cities. You should know: basically your brain is much better than any computer chip. Now it is about achieving as human a driving behavior as possible – with software that is not as strong as the human brain and that, above all when anticipating and “feeling” situations, still struggles a lot.
That is probably harder in urban areas than on the highway?
Oldemeyer: Correct! The German Autobahn is, despite
sometimes high speed differences, quite “learnable” and relatively simple. With this end-to-end driving demo – i.e., from the trade fair grounds to the trade fair grounds via highway, country roads and lots of downtown – we wanted to give a glimpse of what is technically feasible. For this, we had only four weeks in this case, but that was also exciting for us: How far can we get in four weeks?
Already quite far. Many details and critical points the car has already recognized well and implemented accordingly…
Oldemeyer: Thanks to AI you can achieve very rapid progress here in months, as the vehicles generate more and more data for “learning.”
How do you assess the sensing without LiDAR? Is that enough?
Oldemeyer: We believe not in the long run. For example, in 2026 the laws in China will be tightened again. There is a dark tunnel there where standard objects must be detected – and they are getting smaller! For that you need 4D radars or LiDAR. And after we manage to produce LiDAR systems that take up little space, fit between the cameras in front of the windshield, and are 20 to 30 percent cheaper than the competition – with better performance – we firmly believe in the LiDAR system.
And you avoid the “taxi sign” on the roof!
Oldemeyer: You will laugh: In China, customers love this “taxi sign,” because people like to showcase technology and signal to the neighbors that they are using the latest LiDAR technology. There are LiDAR boxes even sold as “fake” in the aftermarket there. In Europe it's different: Here we prefer to keep our technology hidden. It is behind the windscreen as well, especially since it can always be cleaned there by the wipers. There are many details like this that make autonomous driving!
The interview was conducted by Gregor Soller, Editor-in-Chief
