He Xiaopeng Predicts that Robotaxi Will Reach A Turning Point in Two Years

On July 11th, He Xiaopeng, Chairman of XPeng Motors, stated that the current capabilities, regulations, and commercial ecosystem of Robotaxi are not meeting expectations. However, it is currently in an accelerating development stage. He predicts that Robotaxi will reach a turning point in approximately two years.

In addition, He Xiaopeng also stated that many companies currently developing L4 autonomous driving technology still use the traditional algorithm + small AI model technical route. He suggested these enterprises to quickly switch to the ‘end-to-end’ technology.

End-to-End Learning, also known as end-to-end training, refers to the optimization of the overall goal of a task without modular or staged training during the learning process. It emphasizes the entire process from user input to result output, with no need for manual intervention in between.

The XPeng unlimited XNGP function has been launched for public beta testing with the Tianji XOS 5.2.0 Pioneer 1 version. The official claim is that this is the first mass-produced on-board end-to-end AI large model in the country, and XPeng is one of only two global companies to implement end-to-end technology in vehicles.

SEE ALSO: XPeng’s XNGP ADAS Expands Coverage to 191 Cities

At the May conference this year, He Xiaopeng stated that the goal of its end-to-end large model is 1 iteration every 2 days, and the intelligent driving capability will increase by 30 times in 18 months; by the third quarter of 2024, every street in the country will be drivable.