Article by Zheng Wen

Editor｜Roomy

Some people gaze at the stars, some are down-to-earth, some seek the impossible, while others seek new maps, some unite while others reach new heights…

Let the bullet fly a bit longer.

In this hot summer of 2022, the automotive industry is equally thriving, with praise and controversy. But everything represents the unprecedented boiling of the industry. Each participant can tell a different story.

Due to the foresight layout of “new energy” before, BYD has entered the period of brand dividend supported by technology and products. In the first half of this year alone, BYD sold 633,777 vehicles, with a market share of 28.2% in the new energy market.

Xu Zhengwu, Assistant General Manager of BYD Dynasty Online Sales Business Unit and Marketing Director of the Eastern War Zone, shared that with the launch of blade batteries and super hybrid DM-i technology, BYD has become an absolute king among new energy automotive companies with a monthly average sales volume of over 100,000 units, led by the Han Tang series in the Dynasty family.

Li Yunfei, General Manager of BYD Automotive Brand and Public Relations Department, once said that almost all mainstream automobile brands are exploring the possibility of cooperation with BYD on battery supply.

The long-rumored Tesla will use BYD’s “blade battery”, which has recently been confirmed. According to the latest news released by foreign media, BYD’s “blade battery” has been officially approved by the European Union, and the Tesla Model Y will be the first model to use the “blade battery”.

According to the disclosed file information, the new Model Y is called the 005 model, with a BYD battery capacity of 55 kWh and a range of 440 km. In addition, it is reported that BYD’s “blade battery” has been delivered to Tesla’s super factory in Berlin, Germany, and the new car is expected to be launched as early as the end of August or the beginning of September.

These are just appetizers. The industry has developed to the present day, giving the first sweet jujube. More and more companies are still rushing forward for more “possibilities”.

On August 11th, Lei Jun, chairman of Xiaomi, gave his third annual speech, and all the content about the automotive part was concentrated in the last half an hour of “One more thing” during his two-and-a-half-hour speech.

In March last year, Xiaomi announced its entry into the intelligent electric vehicle industry. More than 500 days have passed, and periodic achievements have been presented through an autonomous driving video.It can be seen that besides automatic roundabout navigation, most of the demos displayed are typical scenarios. It’s worth noting that this demo has been edited. In the self-driving industry, there is a default rule on the Internet that self-driving test videos are unedited. The Xiaomi post-editing and rendering marks are obvious, and the final effect is more of a “promotional film.”

However, considering that Xiaomi’s automotive business has not been established for a long time, it has achieved good results.

Lei Jun said that Xiaomi’s automotive business has built a team of over 500 elites, formulated a technical strategy covering core technologies such as perception prediction, high-precision positioning, decision-making planning, and self-developed algorithms with full-stack capabilities, and worked with multiple internal teams at Xiaomi to focus on polishing autonomous driving technology with the goal of entering the first camp of the industry in 2024.

“We have invested 3.3 billion yuan in research and development costs in the field of autonomous driving.” Lei Jun also revealed that Xiaomi’s automotive business will invest more than 140 autonomous driving test cars in the first phase and carry out research and development verification work throughout the country.

Currently, Xiaomi’s own trial production cars have not yet come out, so in order not to delay the research and development of autonomous driving, a batch of test cars from other brands have been modified. We can see from the test video that there are BYD Han and BMW 5 Series.

In addition to disclosing the latest developments of Xiaomi’s automotive business, Lei Jun also released the bionic robot “CyberOne”, which brought humanoid robots to the public before Tesla did.

In fact, mainstream automakers have a strong sense of urgency about the development progress of autonomous driving. However, the future is bright, and the road is tortuous. On the one hand is the bold vision and practice of automakers in autonomous driving, on the other hand is the contradiction between people’s perception that it is “autonomous driving” and the fact that it is only driving assistance in actual market application.

On August 10, a traffic accident that caused casualties occurred on a certain elevated road in Ningbo, Zhejiang, which is suspected to be related to the LCC (lane centering assist) function of XPeng P7’s assisted driving.

The next day, XPeng Motors confirmed that the vehicle involved in the accident belonged to XPeng, and the owner of the vehicle collided with the personnel of the front inspection vehicle, resulting in the death of a passenger in the front vehicle. XPeng stated that it had gone to the scene to assist in handling the incident in the first place.

The owner, Mr. Wang, said: “At the time of the incident, he was using the LCC function, but the function failed to recognize the person and vehicle in front, and there was no warning before the collision occurred. He was distracted while driving in assisted driving mode and failed to respond, resulting in the accident.”L3 is the watershed between the OEM and the driver’s responsibility. The intelligent driving assistance systems equipped in new cars in the current market are all L2 and below levels where the driving responsibility lies with the driver.

XPeng Motors pointed out in the user manual that car owners cannot rely on the ACC/LCC system and need to be prepared to intervene manually at any time. The user manual listed 52 warnings related to LCC/ACC, including scenarios where the system may malfunction, such as failure to detect stationary vehicles and objects. NIO and Tesla have also encountered similar problems.

For example, in the NIO ES8 user manual, the definition of NOP is the deep integration of the navigation system and Pilot’s automatic driving assistance function. However, NIO also clearly states in the user manual that the “Navigation Assistant” Beta version is a public test version, and the function is still in a continuous optimization stage. One of the many warnings listed in the user manual is that “like Pilot and other driving assistance functions, Navigation Assistant cannot respond to static obstacles (such as roadblocks, triangular warning signs, etc.). If there is an accident or construction area ahead, please take control of the vehicle immediately to control direction and speed.”

Nonetheless, accidents still occurred during actual usage.

Does following the OEM’s user manual rules ensure safety? This is not always the case. Tesla is a typical car manufacturer suspected of misleading consumers by excessively promoting “driving assistance.” In fact, since Tesla’s prevalence, many people around us have mistakenly referred to driving assistance functions as “automatic driving.”

The warning bell needs to be sounded for a long time. Recently, the California Department of Motor Vehicles (DMV) accused Tesla of seriously expanding the functions of Autopilot and FSD in its promotional language, suspecting advertising fraud and potentially revoking its license to produce or sell cars in California.

Although Tesla has posted a disclaimer telling users that “active driver supervision is required, and the vehicle cannot drive itself,” this alone cannot exempt Tesla from responsibility.

A somewhat regrettable study found that when humans and machines share control of the steering wheel, if they suddenly require humans to take control of the steering wheel in an emergency, humans may be unable to handle it.A series of events and ringing alarms demand that automakers not only restrain their marketing in the market competition, but also emphasize education on “safe use” among consumers. Therefore, in his annual presentation, Lei Jun repeatedly stressed the adoption of full-stack self-developed algorithms and the full-scale development of autonomous driving technology, hoping that this would be a medium- to long-term layout. However, Xiaomi must prepare in advance for not only investing huge amounts of capital, but also absorbing the cost of trial and error during the whole process.

After all, the progress of autonomous driving technology has no accelerator to press or shortcuts to take, and safety issues have always been of primary importance.

If Xiaomi’s more-than-500-day progress in vehicle manufacturing has not been fast enough, Jidu, which was established one month later, has moved relatively quickly through cooperation with Geely Automobile. On August 8th, Jidu’s first automotive robot ecosystem partner conference was held at its newly completed JIDU ROBO BASE headquarters in Shanghai Jiading.

At the conference, Jidu CEO Xia Yiping introduced in detail to the attending partners the core competitiveness of Jidu’s automotive robot products, which are highly intelligent.

Before this, when Xia Yiping first met Li Shufu, Li threw him a series of questions: “Have you figured out how to prepare the brand? How to sell? How to build a sales system?” Xia Yiping was confused by the series of questions.

More than a year has passed, and Xia Yiping deeply feels that building a car from 0 to 1 is an extremely complex system engineering. The problems of production capacity coordination, test verification, supply chain streamlining, and sales network construction that follow one after another constitute the daunting iceberg. But he also has a deeper understanding of the automotive industry.

He began to propose clear time points. By 2023, Jidu’s first automotive robot product delivered in volume production will have the best “door-to-door, start-to-stop” high-level autonomous driving ability in the industry at that time.

Xia Yiping revealed that Jidu will soon release a limited edition of its first mass-produced automotive robot model and will simultaneously start limited pre-orders. It is expected to officially deliver in the second half of 2023. The appearance design of its second model will be unveiled at the Guangzhou Auto Show at the end of 2022 and is expected to start delivery in 2024.

In addition, Jidu’s first brand experience center will open in Shanghai this year, and the first batch of stores is planned to enter 46 cities in China in 2023, initially completing the national layout of its sales network. Xia Yiping also announced the “2880 Plan,” which means that by 2028, Jidu will have the ability to deliver 800,000 automotive robots annually.

On-site, Baidu founder, chairman and CEO Robin Li and Li Shufu, chairman of Zhejiang Geely Holding Group, expressed their views through video linkages.

Li Yanhong said, “Regarding autonomous driving technology, Jidu will lead ahead of Tesla.” Li Shufu stated that in the future, Geely will continue to increase its R&D support for Jidu’s products, empowering Jidu with Geely’s more than 20 years of global intelligent manufacturing capability, engineering technology R&D capability, industrial chain integration capability, and global collaboration capability.

The support of “Two Lis on the Same Stage” means that the perfect supply chain supported by Geely + Baidu allows Jidu Automobile, the “latecomer to car manufacturing,” to have “efficient” research and development speed.

Not only Jidu Automobile but also the Avita brand jointly created by three major players are advancing side by side with two forces.

Established in May 2021, Avita was later than Xiaomi and Jidu in timing. However, Avita seems to have regained some of the time difference through the “strength in numbers” of its cooperative mode of operation. CHN, a combination of Changan Automobile, Huawei, and CATL, has pooled resources. It is better to say that the CHN platform is not so much ahead as a leading pioneering collaborative model.

On August 8, Avita launched Avita 11, with three models priced between RMB 3.499 million and RMB 4.099 million. The Long Endurance Dual-Motor version and the Long Endurance Dual-Motor Premium version will begin deliveries in December this year, and the Ultra-Long Endurance Dual-Motor Premium version will begin deliveries in the first quarter of next year.

The market Avita is entering is still highly competitive, and its future performance needs to be further observed after delivery.

Looking only at the ideal state, this combination is bound to burst with enormous market energy. However, it is worth noting that in actual cooperation, friction and consumption may be greater than imagined, which will test the team’s long-term cooperation and coordination.

In about 500 days, Xiaomi, Jidu, and Avita present three different market samples, all of which are the industry’s appearance during special times.

Those of us who are in it can already vaguely touch the outline of the future. At this time, the ancient and traditional industry, in addition to patience, may need the most imagination. It is based on the thought experiment of civilization.

Liu Cixin, the author of “The Three-Body Problem,” said, “When artificial intelligence has intelligence exceeding that of humans, imagination may be the only advantage we have over them.”

Mary Shelley’s “Frankenstein,” the earliest recognized science fiction work, “was born” at a time when industrial revolutions were being completed and scientific civilization was soaring in the western countries. At that time, people began to believe in the power of science.The following science fiction books, like imagination with wings, are endless. The seemingly absurd plot described in the book is becoming more and more close to the scientific reality.

The Nautilus in Jules Verne’s “Twenty Thousand Leagues Under the Sea” became a real submarine; the aerial vehicle in “Five Weeks in a Balloon” was upgraded to an airplane; Dr. Number in “Little John” said that everything in the future will become numbers… This is the fascinating part.

“Science fiction arranges various possibilities of the future like a pile of imaginative pebbles, which are there for people to admire and play with,” said Liu Cixin in the preface of “The Three-Body Problem.” Science fiction does not predict the future, but now the automotive industry is also undergoing the transformation of “translating ideals into reality.”

This article is a translation by ChatGPT of a Chinese report from 42HOW. If you have any questions about it, please email bd@42how.com.