How can domestic SEV surpass Tesla?

Author: Xu Jinkai

Tesla is the creator of Smart Electric Vehicles (SEV), but not the leader. This is not determined by the current level of technology, but by differences in the concept of car manufacturing.

As early as 2006, Musk proposed the first chapter of the macro plan, which envisaged the development of affordable cars for the masses. Now, this idea has become a reality.

Recently, Tesla officially announced at its investor day that it will reduce the production cost of the next-generation platform by 50\% and establish a new goal of selling 20 million cars per year.

According to the China Association of Automobile Manufacturers data, the overall sales of China’s passenger car market in 2021 was 20.146 million, and it is expected to reach 20.543 million in 2022. The goal of Tesla’s annual sales of 20 million cars is almost equal to the entire Chinese market.

From the perspective of industry development, Tesla, which continuously reduces costs and seeks greater market opportunities, is the promoter of SEV’s popularization. However, popularity means being ordinary, and Tesla’s originally high-end brand image will gradually disappear.

Unlike Musk’s idea of making Tesla affordable to everyone, SEV is an opportunity for domestic brands to achieve a breakthrough and form a brand advantage in the high-end market. From “NIO Xuebao” to Xpeng, Landtour, IM, and Avatar, all of them are rooted in the high-end SEV market. Domestic SEV brands have also started a comprehensive competition from products to services. If you want to achieve a breakthrough in the extremely competitive market, you must have advantages.

Therefore, NIO has targeted services, Ideal has targeted new fathers, XPeng has focused on technology, Landtour emphasizes safety, IM is chatting about driving, and Avatar highlights intelligent driving.

The high-end technology brand image of Tesla is largely contributed by FSD intelligent driving assistance, but currently, Tesla’s driving assistance function is not advanced, and Avatar, supported by Huawei, has already surpassed Tesla.

Long-term Negative View on Tesla

“Avatar and XPeng will launch their city-level NOA functions this year, and more and more brands in the Chinese market can replace Tesla.”

On the one hand, domestic SEV brands are developing rapidly, and on the other hand, Tesla, which has been plagued by “out of control events”, is also experiencing a gradual collapse of its brand image.“`

In addition to its frequent price cuts, Tesla has not only hurt the brand obsession of old car owners, but also lowered the entry barrier, quickly dissipating the high-end brand image it used to have.

However, Tesla’s difficulties are far from over.

For many consumers, their first encounter with Tesla was through its intelligent driving capabilities. Videos of Tesla vehicles on overseas streets “driving themselves” with the FSD function may have aroused many people’s “worship of foreign things” psychology.

However, in reality, Tesla’s advantage in intelligent driving capabilities is not obvious.

Firstly, there are too few car owners subscribed to FSD.

According to David Lau, vice president of software engineering at Tesla, approximately 400,000 users have subscribed to FSD, while Tesla’s current global inventory is approximately 4 million vehicles. Therefore, the subscription rate of FSD is about 10%, and this rate is even lower in the Chinese market.

Secondly, Tesla’s position in terms of assisted driving functionality is not absolutely leading, as of now.

Hou Cong, co-founder and CTO of Light-boat Wisdom Navigation, stated at the 2022 Heart of Car’s “2022 Smart Car Technology and Business Innovation Forum” at the end of 2022:

“Tesla’s pure visual solution belongs to the mid-range in terms of hardware configuration, but it can achieve high-level effects in relatively standardized traffic conditions in North America. However, China’s road conditions are more complex. The roads are not only irregular, but driving is also irregular, and pure visual detection has some defects in the recognition of obstacles. These issues can be well addressed by coupling with LIDAR.”

Tesla’s assisted driving functions can greatly reduce driver fatigue on highways and city expressways where the scenarios are relatively simple. However, its assisted driving functions are no longer useful in complex urban scenarios.

Domestic SEV brands represented by Avita and XPeng have shown a trend of surpassing Tesla in the urban NCA field. On March 9th, Avita officially announced that it has opened up Avita 11 city NCA smart driving navigation assisted driving functions for user test drives in Shanghai and Shenzhen.

Meanwhile, the two cities of Guangzhou and Chongqing are also about to open for experience.

On one hand, Avita is gradually pushing its NCA in the urban area towards consumers, while on the other hand, the long-rumoured latest hardware HW4.0 for Tesla's self-driving is absent from the investor day. This has largely indicated that, in the road towards intelligent driving, domestic SEV brands have the strength to catch up with Tesla.

As a result, Tesla's advanced brand aura will inevitably fade, gradually becoming an ordinary brand that needs to sharpen its competitiveness in terms of "cost-effectiveness".

## Tesla Goes Left, Avita Goes Right

On the surface, Avita's catching-up with Tesla in the field of intelligent driving is the result of technological accumulation, but fundamentally, this is due to different car-making concepts between the two parties.


Tesla aims to be a popularizer of SEVs, which determines that it must pursue cost-effectiveness more, reduce prices or launch lower-end models to seize more market share.

Avita, however, is committed to creating a high-end SEV brand, more focused on safety and experience. These completely different car-making concepts determine the differences between the products of the two sides, which has also been confirmed in the recent media comparison of the two.

Recently, a media comparison of Avita 11 and Tesla Model Y, which are positioned similarly, found that in terms of design and manufacture, Tesla uses more "compromise" ways to meet user needs;

While Avita focuses on the foundation and material. Taking intelligent driving-related configurations as an example, the two have obvious differences in relevant configurations of intelligent driving.

Avita 11 is equipped with 13 cameras, 3 lidars, 6 millimeter-wave radars, and 12 ultrasonic radars, with L3-level assisted driving capability in advance;

Model Y is equipped with 8 cameras, 1 millimeter-wave radar, and 12 ultrasonic radars, enabling L2+ level assisted driving.

Without discussing the not yet widely deployed urban NCA function, at present, when the regulations of automatic driving at the L3 level are not yet fully allowed, the performance of the assisted driving functions of Avita 11 and Model Y in simple scenarios such as highways and city expressways are quite similar, which is the common perception of most consumers.

While achieving similar driving assistance effects, Model Y uses fewer sensors and lower computing power than its counterpart Autowise 11. This state indicates two obvious issues:

Firstly, Tesla’s “technological leadership” aim to reduce costs;

Secondly, Autowise, which leads in hardware, has a higher bottom limit than Tesla when the driving assistance effect limit is similar.

However, it is worth noting that the intelligent driving industry is developing rapidly, and the widespread application of NCA in urban areas is an inevitable result of industry development this year. At the same time, first-tier cities led by “Beijing, Shanghai, Guangzhou, and Shenzhen” have gradually begun to explore L3-level driving assistance, and higher-level driving assistance functions are emerging.

It won’t be long before Model Y’s advantages in driving assistance will become a disadvantage. At least in terms of hardware, it will be difficult for Model 3 and Model Y to achieve a leap to higher-level driving assistance.

Autowise 11, which has strong hardware support, can upgrade its driving assistance function through OTA. This is the confidence behind its claim to have L3+ level driving assistance functions. Judging from the current industry consensus that L3-level driving assistance functions still need a long time to accumulate data, Autowise 11’s driving assistance ability will be at a more leading level for a long time in the future.

Tesla, which frequently lowers its prices, has not only “backstabbed” old car owners in terms of model prices but also seems to harvest old car owners’ suspicions of driving assistance functions. The fundamental reason for this result is that Tesla has strict cost control.

Thus, Tesla, which advocates “scale wins,” and Autowise, which emphasizes “quality wins,” have taken completely different paths. Although the functional difference is not significant at present, Autowise’s advantages will quickly show up next.

Do your homework and exceed Tesla with experience

From the perspective of SEV industry development, Tesla resembles the smartest student in the class, who occasionally comes up with many new ideas. Its new problem-solving ideas are clearer and can directly solve problems.


BEV+Transformer technology route owes its success to the contribution of Tesla.

Moreover, Tesla has adopted a more integrated electronic architecture, a three-electric system, and a vehicle structure. From this perspective, Tesla has become a learning model for many domestic SEV brands.

However, Tesla’s “smart moves” do not prevent its classmates from advancing and catching up steadily. Avita’s multiple layouts in the field of intelligent driving have laid the foundation for it to gain a leading advantage in the upcoming city NCA industry environment.

According to the latest global autonomous driving company competitiveness ranking released by consulting firm Guidehouse, Mobileye, Waymo, and Baidu are the top three, while Tesla has remained ranked last for three consecutive years.

“Too many marketing promises, serious lack of fulfillment, and insufficient safety level.” This is the main reason for Guidehouse’s extremely low evaluation of Tesla.

It is worth mentioning that in the context of last year’s shortage of chips in the automotive industry, the Chinese-made Tesla Model Y and Model 3 canceled redundant steering components. Tesla stated that the cancellation of steering redundancy affects L3-level autonomous driving capabilities, but the market is currently in the L2 stage, so this approach will not affect L2 assisted driving.

In contrast to Tesla’s approach of reducing features, Avita 11 has made sufficient preparations for the industry’s pursuit of dual redundancy in the L3 autonomous driving field.

According to media teardown results, Avita 11 is equipped with three redundant systems for braking, steering, and power supply. Taking the braking system as an example:

Avita 11 is equipped with two braking systems from Mainland China’s MKC1 and HBE. During normal driving, MKC1 can provide normal braking force to all four wheels. However, once the MKC1 system fails, HBE will quickly take over the braking to provide 0.6g of braking force to the front wheels.From the hardware assembly of 34 sensors of the assisted driving system and three complete redundant systems, the Aion LX (AVATR 11) already has the physical foundation to achieve L3 assisted driving functions. As soon as regulations allow L3 level assisted driving vehicles on the road, Aion LX (AVATR 11) can quickly upgrade the relevant functions through OTA, and in the field of intelligent driving, it has leading and superior technical capabilities.

In addition to the comprehensive layout in the field of assisted driving, the advantages of Aion LX (AVATR 11) in the use experience are also quite obvious.

If the cast aluminum chassis parts and ubiquitous sound insulation cotton are the normal operations that the Aion LX (AVATR 11) use to exceed Model Y, then ASE active noise, off-board cabin sensing, and RNC active noise reduction system and other functions are the key advantages that enable Aion LX (AVATR 11) to crush Model Y with high-quality experience.

Take the RNC active noise reduction system as an example. Unlike ANC engine active noise reduction in the traditional automobile field, the difficulty of RNC road active noise reduction is higher. After all, the road noise produced under different road surfaces, different tires, and different temperature conditions are different, and it is obviously more difficult for RNC than ANC.

Aion LX (AVATR 11) still maintains its steady style. By disassembling the vehicle, it is discovered that Aion LX (AVATR 11) is equipped with four vibration sensors and four noise reduction microphones, and the vibration sensors are responsible for collecting the vibration causing road noise. The in-car microphones collect real-time road noise in the car, and use it as the noise reduction target. Then, through the car’s DSP (digital signal processor) to run the noise reduction algorithm program, generate equal-sized opposite noise, and achieve the purpose of reducing noise.

It is not difficult to imagine that such a solid and user-centric style of making cars, coupled with the AVATR Care service, can provide Aion LX (AVATR 11) consumers with a leading user experience from product functions to service levels. Although this steady and firm approach is time-consuming and laborious, it has a good effect in ensuring the level of consumer experience.

Perhaps Tesla, which pursues product excellence, will never understand that domestic SEV brands represented by Aion LX (AVATR 11) can replace itself by pursuing quality and become the leader in high-end intelligent electric Vehicles in the new era.

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