Logged-out Icon

China’s XPeng gearing up to launch a robotaxi network with G9 SUV

Xpeng

XPeng, a Chinese luxury electric vehicle startup is gearing up to launch its robotaxi fleet. The company launched its latest G9 SUV in September this year, which is China’s first mass-produced commercial vehicle to pass a government-led closed field autonomous driving test. The G9 will be used for XPENG’s robotaxi fleet, the company said at its fourth annual 1024 Tech Day on Monday.

XPeng

During the launch, XPeng said the G9  would be equipped with the company’s new Advanced Driver Assistance System (ADAS), the XNGP, which combines XPeng’s Highway Navigated Guided Pilot (NGP) and City NGP to perform specific driving functions to automate both highways and urban driving scenarios. The XNGP is good enough to lay the groundwork for a robotaxi network, and the G9 can help scale that network, says XPeng’s Vice President of Autonomous Driving, Dr. Xinzhou Wu.

“Obtaining road test approval for our series-produced commercial vehicles – without retrofitting – is a great achievement,” Wu said at XPeng’s Tech Day, further adding, “our platform-based Robotaxi development aims to generate significant cost advantages and ensure product quality, safety and user experience.”

XPeng unveiled XNet neural network-based architecture at Tech Day. The company says XNet varies from its first-generation visual perception architecture by adopting a deep neural network that was developed in-house to deliver visual recognition with human-like decision-making capabilities, drawing data from multiple cameras.

The company explained that its autonomous driving technical stack can reach 600 PFLOPS, increasing the training efficiency of the autonomous driving model by over 600 times. To that note, model training can be significantly reduced from 276 days to just 11 hours.

These upgrades have enabled XPeng to establish an entirely closed-loop autonomous driving data system (data collection, labeling, training, and deployment) that utilizes lightning-fast machine learning to consistently self-improve, according to Xpeng’s press release.

This website uses cookies to ensure you get the best experience on our website