With the end-to-end model installed, Xpeng Motors' intelligent driving is upgraded again. Will it become a panacea for sales?

Richard Yu, Chairman of Huawei's Automotive BU, has always been a big shot on the Internet, nicknamed "Big Mouth Yu". So has Richard Yu ever been defeated by anyone?

There are some. At the 2024 China Electric Vehicle Hundred People Forum, Yu Chengdong publicly stated that in the past, Xiaopeng was the best in China in terms of smart parking and smart driving.

However, in the past few months, Xiaopeng's advantage in the field of intelligent driving seems to be disappearing. The ADS 2.0 advanced intelligent driving assistance system launched by Huawei has occupied the technological high ground and successfully contributed to the popularity of models such as M7 and M9. At the same time, BYD, which has always had no voice in the field of intelligent driving, has officially started the second half of its intelligentization and launched the Eye of God intelligent driving assistance system.

Xpeng Motors must fight back because intelligent driving is the biggest card in its hand.

On May 20, Xpeng Motors held an AI DAY press conference and officially announced the end-to-end large model. According to Xpeng Motors officials, it will achieve full map-free driving and be able to drive on every road in the country in the third quarter of 2024, and achieve L4 level intelligent driving in 2025.

End-to-end intelligent driving assistance has been a hot topic in the field of intelligent driving in recent months. The concept was first proposed by Tesla's FSD system. During the previous Beijing Auto Show, Huawei released the ADS3.0 intelligent driving system and also launched "end-to-end" intelligent driving assistance functions. However, Xiaopeng Motors' end-to-end large model has achieved "the first mass production and listing in China". In the field of intelligent driving, Xiaopeng Motors has regained the dominant position.

What is end-to-end driver assistance functionality?

In fact, the principle is not complicated. The previous intelligent driving principle is as follows: the on-board perception network perceives the road conditions during the vehicle's driving, and then compares them with manually written rules. If they match, the vehicle will be controlled. If the match fails, the control of the vehicle will be abandoned. At this time, a human driver needs to take over.

This is also the main reason why many manufacturers like to promote their "zero takeover within XX kilometers". The fewer the takeovers, the higher the matching degree and the more advanced the intelligent driving level. However, manually written rules always have a lag. When encountering some complex scenes, the perceived content does not match the manually written rules, and the intelligent driving will get stuck. In the previous intelligent driving tests of many brands, vehicles have appeared "stuck in place" at complex intersections.

End-to-end intelligent driving integrates AI big model technology into the execution level. It does not require manually written rules, but can find patterns from perceived data and execute them. The decision-making efficiency is an order of magnitude higher than the current intelligent driving, becoming "more human-like".

This is exactly where Xiaopeng Motors has made a breakthrough this time. According to He Xiaopeng, Xiaopeng Motors will improve its intelligent driving capabilities by 30 times within 18 months, and will iterate its intelligent driving model every two days. At the press conference in April, Huawei proposed a goal of "iterating once every five days", which shows that Xiaopeng does have its own unique skills in the field of intelligent driving.

It should be said that in the direction of intelligent driving, Xiaopeng Motors has always been in the first echelon. The question now is, with higher-level intelligent driving, can Xiaopeng Motors soar in sales?

Let's first look at Xpeng's market performance in the past two years. In 2022, Xpeng Motors delivered a total of 120,000 vehicles, while the sales target for that year was 250,000 vehicles; in 2023, Xpeng Motors delivered 141,600 vehicles. In 2023, when new energy vehicles were booming, Xpeng only had an increase of just over 20,000 vehicles.

At the same time, Xpeng Motors' gross profit margin fell from 11.5% in 2022 to 1.5% in 2023. In the first quarter of 2024, Xpeng Motors suffered a net loss of 1.37 billion yuan. In April 2024, Xpeng Motors' cumulative sales were once again suppressed below 10,000 units, and the monthly sales of the highest-selling Xpeng P7 were only 2,383 units.

Is the decline of Xpeng Motors due to the lag of intelligent driving? No, it has already failed to achieve its sales target in 2022. In the Chinese new energy market in 2022, intelligent driving is not the focus of consumers. People are more concerned about indicators such as space, power, endurance, and comfort.

In other words, the main reason for the decline in sales of Xpeng Motors is the problem of its overall product strength, not because it fell behind in intelligent driving. With Yu Chengdong's arrogance, he publicly admitted until March 2024 that Xpeng's intelligent driving was the best in China in the past. It can also be seen from the side that intelligent driving was not the main focus of users at that time. Of course, the title of the best intelligent driving did not bring corresponding sales to Xpeng Motors.

The question is, if the end-to-end intelligent driving technology released by Xiaopeng is widely used in vehicles and really brings a better intelligent driving experience, can Xiaopeng achieve a breakthrough in sales? The reality is probably not so optimistic. Since Xiaopeng's sales decline is not due to the lag of intelligent driving, its hope of increasing sales cannot be pinned on the upgrade of intelligent driving.

In fact, there is no definitive answer to how important intelligent driving is to a new energy vehicle. The reason why new forces such as Wenjie, Weilai, and Xiaopeng are desperately developing intelligent driving is that they only have intelligent driving that can surpass traditional manufacturers, which is a single-point breakthrough weapon.

In fact, BYD and Geely, which currently have the best sales performance, are not very outstanding in the field of intelligent driving, but their sales have been rising. BYD Chairman Wang Chuanfu once said: The future of intelligent driving is just an ordinary auto part worth three to five thousand yuan, which cannot determine the overall performance of a car.

It is a good thing that Xiaopeng Motors can make breakthroughs in intelligent driving, but Xiaopeng Motors should focus more on the performance of the whole vehicle. Most consumers are ordinary people. They don’t necessarily need any intelligent driving, nor do they have any need to experience the machine taking over the car. But if a car is particularly outstanding in terms of space, power, endurance, appearance, etc., they will immediately vote in favor.

Smart driving is now very popular, and anyone who doesn't try it out will feel embarrassed to go out. But from the perspective of car users, they want more issues that are closely related to their daily car use. Smart driving is important, but it is not the main reason for consumers to open their wallets.

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.