One electric motor per wheel gives the possibility for some really awesome torque vectoring. I could see a machine learning model being used to figure out the torque to apply to each wheel given the feature inputs of turn angle, lateral g, suspension height, tire temp, road temp, humidity, etc.

Or you could just solve it analytically, making validation easier.

With fewer environmenal variables this is already done, to a degree, for locomotive-wheel traction motors. For example pulsing the torque in frequency and amplitude to manage wheel-slip on wet rails.

The website doesn't really give a context, but I found this article to be helpful: http://www.wired.co.uk/article/nextev-hypercar-nio-ep9

Thanks, we changed to that one a while ago. The submitted url was http://www.nio.io/ and the submitted title was 'NIO EP9 “Fastest Electric Car in the World”'.

Why does this site have Baidu ads? It IS an ad.

"Each car costs approximately $1.2 million to build."

So, $1.2m and still lags behind the model s for 0-60mph by a wide margin?

I'm pretty sure Model S can't complete even half of the Nordschliefe without going into limp mode, so the comparison is not really fair.

0.2s is a wide margin?

Model S took way more than 6 Million dollars for the design.

What about 0-100? 60-100? 100-150?