Hans-Peter Hannighofer put on his helmet, pulled down his visor and eased himself into a bobsled, ready to tackle the Beijing 2022 course. Hannighofer, however, was nowhere near the Chinese capital. And the track facing him was not made of ice, but rather a projection onto a 270-degree cinema screen. 

For the past 18 months, Germany's bobsled team have been plotting their assault on the Beijing medals table in a nondescript office building in the north of Munich. Together with the German carmaker BMW, they have created a simulator, modeled on motorsport technology, that they hope will give them an edge over their rivals at the Winter Olympics. 

"The simulator is very realistic, especially visually," Hannighofer told DW during a recent demonstration. "I wouldn't have imagined it could be as good as it actually is. It really feels like a real-life bobsled." 

That level of realism is achieved by mounting the bobsled onto a movable platform normally reserved for racing cars. A motor connected to the platform replicates the bumps and collisions that a bobsled pilot experiences on the ice. The pilot steers the bobsled as normal using rope pulleys, with another motor at the front generating torque in the turns. 

Because it was built for the Olympics, the Beijing course is largely unknown in comparison with other tracks around the world. Add to that the fact that opportunities to practice on it have been severely limited for athletes outside of China due to the coronavirus, and the advantages for Germany — long used to success in bobsled — are clear. 

"It helps us tremendously," said Richard Oelsner, a backup pilot for the Germany team. "Here we can rework, readjust and continue to develop a feeling for the track. Without this, it would definitely be more difficult to get everything right."  

Hannighofer, who missed out on a place at the Olympics through injury, echoes that sentiment: "It's a huge advantage to be able to come here and get an impression of the track. It means you're a step ahead when you get there for real." 

Fine-tuning the performance 

In a separate control room filled with screens displaying telemetry data and a computer-generated image of the bobsled, Julian von Schleinitz watched Hannighofer and Oelsner barrel virtually down the track. Von Schleinitz — a junior world champion in another sliding sport, luge, is the brains behind the project. He says his own experience of competing inspired the idea for a bobsled simulator. 

"I was always thinking it's so cool seeing computer video games, like in motorsport, where you can just try things in a simulation," von Schleinitz said. "And then, when you are really good at it, you go on the real track in the real world. I always wanted to have this." 

Initially, the track for the simulator was based on the one at Königssee in southern Germany. But as more information became available about the Beijing course, the software was updated to more closely reflect what awaits the team later this month. 

It is the software and not the hardware (the movable platform and motors) that brings with it the most difficulty, according to von Schleinitz. But getting that right reaps the biggest rewards: the amount of data generated means the simulator doesn't just allow the pilots to get used to the Beijing course, but also to analyze how and where on the track they can improve their performance.  

"There's a lot of data we gather in the simulation," von Schleinitz said. "We see what is happening with the bob, if the bob is sliding or if it's gliding straight.  

"We also see what the athlete does, which is very important for us," he said. "We see the steering actions, and we see all the driving lines in three dimensions. So we really have a complete picture of what is going on in the simulation." 

In turn, firsthand knowledge from the pilots can also be fed back into the software to fine-tune the simulator's realism. That has included smoothing out some of the harder-to-create corners and making the bobsled less sensitive when it clips the edge of the track.  

A course like no other 

Oelsner, one of the pilots who took part in a bobsled test event in Beijing in October, said the Olympic course had "very unique" characteristics. 

"You really have to experience it to understand it," he said. "The track is particularly long. It has a few special passages and combinations that you don't find anywhere else in the world. It's a very fine line between fast and slow, between winning and losing everything." 

Staying on the right side of that fine line will be key. "There's virtually no margin for error," Olympian Johannes Lochner told DW at a World Cup event in Germany's Winterberg last month. "You have to be within a few centimeters of the curves on the upper part of the track. If you make a mistake there, all the speed is gone, and you lose time to the finish."  

Lochner, who's taking part in his second Olympics, also trained on the simulator. 

"It gave us a good feeling for the track," he said. "You can mentally prepare for it. You always have to have a plan B. If you make a mistake, what do you have to do to get back on the right line? Do you already have these preparations in your head? We've done all that before we get to Beijing." 

Germany is an established force in bobsled, having won all three gold medals on offer at the 2018 Winter Olympics (one was shared with Canada) and investing large sums of money in the sport. 

Yet Oelsner rejects suggestions that the simulator, which BMW says is likely the only one of its kind the world, is further unfair on Germany's rivals, pointing out that Chinese athletes have had "hundreds or even thousands" of goes on the track, compared with about 40 for other competitors. 

"If you have it, why not use it?" Oelsner said. "We want to get better and win the Olympics. The Chinese have been using the track since last year. They have a bigger advantage than us. We're just trying to keep up as best as possible and do everything we can to even the playing field so that we come away with the gold medal in our hands." 

DW's Davis VanOpdorp contributed to this report. 

Edited by: Matt Pearson.