When driving at night in China, large trucks with broken taillights are likely to be seen, posing a challenge to camera recognition, Li Auto's CEO said.

(A Tesla Model 3 on display at the June 2024 new energy vehicle show in Shanghai. Image credit: CnEVPost)

The debate over whether smart electric vehicles (EVs) should carry LiDAR continues, with the CEO of Li Auto defending the use of the component in his company's models.

“I believe that if Musk had ever driven on different highways in China deep in the night, he would have chosen to keep a LiDAR in the front as well,” Li Xiang, founder, chairman, and CEO of Li Auto, said at an AI Talk event on Wednesday, according to a text released yesterday by the automaker.

Li mentioned this in response to a question about why Li Auto is using LiDAR when Elon Musk' Tesla isn't.

Tesla takes safety just as seriously, but Musk needs to understand the driving environment in China, Li said.

Li Auto is reserving its use of LiDAR, not because the technology is bad, but because of safety, Li said.

“China is different from the US. If you regularly drive at night in China, you'll see large trucks with broken taillights, and the large trucks with broken taillights maybe even parked right on the main road,” the Li Auto CEO said.

Currently, in an unlit environment deep in the night, cameras can only see objects slightly more than 100 meters away at best, but LiDAR has a detection range of 200 meters, he said.

It is with the support of this capability of LiDAR that Li Auto is able to implement the AEB (automatic emergency braking) function at speeds of 130 kilometers per hour, Li said.

“I think it's very important because our cars are family-oriented and the safety of everyone's life is very important,” he said.

“That's the fundamental reason we continue to keep LiDAR and will still keep it in future models,” Li added.

Tesla is the foremost proponent of the pure vision smart driving solution, and its management had expressed its dislike for LiDAR several times over the past few years.

On December 10, Tesla's vice president of external affairs Grace Tao reiterated Musk's judgment on the autonomous driving route in a Weibo post, saying that only pure vision will enable safer and smarter fully autonomous driving.

Roads and traffic regulations are designed for human eyes, optic nerves and brains. In such roads, only cameras, visual neural networks and autonomous driving hardware can mimic human observation, perception and decision-making habits, making it possible to achieve driving results that rival or even surpass those of humans, Tao said.

If the information from the radar and the camera contradicts each other, the vehicle's brain will have a hard time making a decision, Tao said at the time, adding that the addition of LiDAR will also increase the cost of the vehicle.

In China, LiDAR is now used on basically all mainstream high-end EVs.

Interestingly, Xpeng (NYSE: XPEV) ditched the use of LiDAR for its new sedan, the P7+, which was launched on November 7, after CnEVPost mentioned the shift in strategy in a July 9 exclusive report.

Earlier this month, Xpeng made regulatory filings for the facelifted G6 and G9, with specs hinting at the removal of the LiDAR option for both SUV (sport utility vehicle) models.