Shulou(Shulou.com)11/24 Report--

In the last two months, two major events have taken place in the lidar market:

For one thing, ibeo, a lidar company, filed for bankruptcy because it was unable to get more financing.

The ancestral German lidar company, founded in 1998, worked with Valeo as a technical support to help Audi A8 launch the world's first lidar production car in 2017.

When Great Wall announced plans to upgrade its smart driving strategy in 2020, it was also enthusiastic to introduce that the WEY brand L3 self-driving model would be equipped with three ibeo all-solid-state lidars.

The other is that Velodyne, once the leading player of in-car lidar, announced a merger with another lidar startup Ouster.

The combined market capitalization of the two companies is about $400 million, but before that, the shares of the two companies have fallen by more than 80%. It is said that the actual situation of the merger is that Ouster acquired Velodyne at a low price.

Five years ago, Velodyne founder David Hall (David Hall) shouted: "We are the leading loners, the finish line is right in front of us; turn around, there are hundreds of angry riders chasing you breathlessly."

Hall was very optimistic at the time. He laughed and said that he only looked forward, and the scenery ahead was better.

But today, five years later, Velodyne apparently didn't have a better view and was acquired at a low price after a total loss of $7.1 million. Holborn was also ousted by the board of directors in February 2021.

The "breathless" pursuers in Hall's mouth ran one after another from the place where their predecessors fell.

Among the pursuers, the rise of Chinese lidar companies is particularly noticeable.

According to the 2022 Lidar Application report in Automotive and Industry released by Yole, two Chinese suppliers of Wesai Technology and Suitengchuang climbed rapidly in the 2021 revenue rankings-Wesai Technology rose to 6th place from 12th last year, and Suiteng Juchuang rose from 13th to 10th place. By contrast, Velodyne dropped from 7th to 13th.

If the 2022 report on the Application of Lidar in Automotive and Industry only looks at the automotive field, Wesai Technology has even become the number one lidar company in the world in total revenue (the top five companies are mainly suppliers in the industrial and surveying and mapping fields).

Some people say that car lidar is another repetitive story for Chinese enterprises that are accustomed to copying the best in the past and will have a high probability of success.

That's true. But Wesai's ability to overtake in the Chinese market, the "front line" of self-driving car competition, has once again proved the irrefutable jungle law of survival of the fittest.

NO.1 [the rise and fall of the Big Brothers] at the beginning of every story, there is a legend. Hall is a legend that has to be mentioned in the field of vehicle lidar.

Li Yimin, CTO of Tudatong, once said that if it were not for Hall, the development of the global self-driving industry would have been delayed by 10 to 25 years.

Hall's establishment of his industrial status began in the stage of local war between the United States and the Middle East.

Before 2004, in order to reduce the number of US military casualties in the Middle East, the US Department of Defense pinned its hopes on the mass production of self-driving vehicles. As a result, a self-driving challenge with a prize of up to $1 million was held.

At that time, David Hall was just the stereo boss. With the steady development of the audio business, the manual "folk inventor" became bored gradually, so he decided to take part in the competition of the Ministry of Defense.

The first year of the challenge was a bit boring and none of the teams even finished the race. In the second year, Hall's participation brought life to the competition.

Hall helped some of the teams to design a roof device like a small flying saucer, which kept spinning and the shaft was buzzing and eye-catching. However, these teams did not win the championship, which was won by Stanford University, which relied mainly on camera sensors and excellent computer vision.

By the third session of 2007, Hall's "Roof Frisbee" showed a greater advantage. Of the six teams that finished the race, five were loaded with "novelty" invented by Hall, including the team that won the championship and runner-up.

This is the prototype of "large flowerpot" (mechanical lidar), one of the main products of Velodyne in the future.

This challenge of the US Department of Defense can be said to be the starting point of the budding of self-driving technology. In addition to lidar, it also inspired the world's first self-driving project.

Google founder Larry Page was the first to act after watching the game. He invited Sebastian Thrun, a Stanford AI Lab professor who represented Stanford, to join Google. In 2008, Thrun took an academic vacation and joined Google with a team from the lab, starting the Google unmanned car project.

When the project became commercially independent, it became known as the self-driving company Waymo. Over the next few years, Cruise, Zoox, Argo, Nuro. A large number of autopilot companies were born in Silicon Valley.

Among Chinese companies, Baidu began to deploy autopilot at Baidu's American research center in 2013, and a number of companies such as Drive.ai, Horse Zhixing, Heduo Technology and Wenyuan Zhixing also became rich in American families in the next two or three years.

The birth of these autopilot software companies also holds up Velodyne's position as the big brother in the lidar industry.

Before 2017, Velodyne basically monopolized the lidar supply for high-performance vehicle lidars, especially for Robotaxi test vehicles.

Even if the whole mechanical parts need to be manually adjusted, time-consuming and labor-consuming, the HDL-64 quotation is as high as $80,000, and the production is slow and expensive, but this does not affect the shortage of HDL-64.

Buyers even lined up at the gate of Velodyne, just to sign the bill, just like today's car companies looking for a core.

Some people joke that how many new test cars autopilot adds each year depends on how many radars Velodyne produces.

Baidu has been a strategic investor in Velodyne since 2016. In 2020, the two sides also signed a three-year sales agreement to achieve a deeper cooperation bundle.

However, the weight of Velodyne in the industry has been gradually diluted with the surge of autopilot.

After 2016, dozens of startups poured in.

First of all, the status of a French company named Valeo climbed rapidly after it received orders from European customers. Subsequently, the list of participants in the lidar field in Europe also appeared on the list of participants in the lidar field in Europe, while Ibeo; in the United States was dominated by Luminar, followed by Quanergy, Innoviz, and so on. Among Chinese companies, Wesai Technology, Suiteng Juchuang, Tudatong, etc., also entered the venue.

On the one hand, the lidar track began to roll in, and more technical routes and technical solutions appeared, on the other hand, the main customers of in-car lidar expanded from Robotaxi to mass production cars, and customers began to re-evaluate the most cost-effective landing solutions.

Ford was one of the first giant companies to invest in Velodyne. However, after making great progress in the research and development of another lidar subsidiary, Ford "liquidated" its entire stake in Velodyne in 2021.

Last year, French parts giant Valeo accounted for 28% of the global lidar market, with Chinese companies such as Suiteng Juchuang, Shevo, Huawei, Wesai Technology and Tudateng having a market share of 10%, 7%, 3%, 3%, 3%, respectively, in the global lidar market. As mentioned earlier, in terms of revenue, Wesai Technology is NO.1.

NO.2 [Chinese companies overtake "three elements]" the resolution of the Velodyne is particularly high, such as 64line, 128line, 256line. But the domestic lidar starts from the low end. " Said Zhang Xiang, dean of the New Energy Automobile Technology Research Institute of Jiangxi New Energy Science and Technology Vocational College.

Cost-effective is the only way for technology followers in many fields to overtake. The same is true of domestic lidar companies.

For the lidar with 360 °rotary and one-dimensional rotating mirror architecture, there are several sets of laser transceiver modules, and there are several lines in the vertical direction, which is called the number of lines. Generally speaking, the higher the number of lines, the more the number of lasers, and the more powerful the comprehensive performance of lidar.

Wesai launched the 40-line lidar Pandar 40 in 2017, and after several iterative upgrades, the Pandar 40p was unveiled at the 2019 CES (International Consumer Electronics Show) for less than $40,000, half the price of Velodyne's "big flowerpot".

With the "fewer lines, lower price" cut from the low-end market, domestic, Wesai won the Velodyne world's largest customer Baidu Apollo (Baidu self-driving business). Internationally, American self-driving companies such as Nuro, Aurora, Cruise and Zoox have also installed Wesai's lidar one after another.

Suiteng launched 16-line lidar mass production in 2017, and at the Beijing Auto Show in October 2018, Suiteng said that they have some foreign products, and they are at least half cheaper.

At present, Suiteng has reached a fixed-point cooperation with Yingchen Technology, Yuanrong Qixing, Momenta, Xiaoma Zhixing, BAIC, Lutes, Polar Krypton, Xiaopeng and other enterprises.

In 2021, the price of the mainstream lidar on the global market basically dropped to less than 2000 US dollars, of which Valeo's latest mass production product Scala 2 sells for 900euros, and the unit price of InnovizOne is about 1000 US dollars, but Chinese companies are even lower. For example, the unit price of Suiteng products is basically between 500 US dollars and 1000 US dollars.

The second reason why domestic companies are able to stand in the head position is that the transformation is fast.

In the past two years, the main market of vehicle lidar has gradually moved from Robotaxi to passenger cars. As the latter requires mass production, both hardware and software need to be certified by car regulations.

However, the original mechanical rotary lidar has defects in volume and life due to the use of a large number of mechanical moving parts, so it is impossible to pass the vehicle specification certification.

In this case, solid-state / semi-solid lidar has become a more promising route. However, Velodyne, which started as a mechanical lidar, has failed to catch up with this technological shift.

In 2021, Velodyne shipped 15000, and the main product is still mechanical lidar. In fact, Velodyne also saw the trend of solid-state and hybrid solid-state technology early on, and MEMS hybrid solid-state lidar was launched in 2017, but Velodyne was unwilling to bear the cost of vehicle specification testing, so it missed a lot of orders from passenger car companies.

At present, Hyundai is the customer of Velodyne's mass production project, but the delivery time is around 2025.

Compared with domestic players, Velodyne obviously lags behind too much.

In 2017, after the Israeli company Innoviz launched the first MEMS hybrid solid-state lidar Innoviz One, Suiteng followed, and the M1 similar to Innoviz One was launched in the same year.

The Xiaopeng G9 uses M1. The polar fox Alpha S Huawei HI version originally planned to carry Huawei's 96-line lidar, but later temporarily replaced it with Suiteng M1 for various reasons.

At the same time, Wosai Technology and Tudatong also quickly completed the route change. The ideal new car L9 is equipped with the semi-solid AT128; provided by Hosai Technology, the ET5, ET7 and ES7 models use the Falcon semi-solid lidar produced by Tudatong.

In recent months, Suiteng Juchuang and Wesai Technology have successively released their own all-solid-state lidar.

In August, Suiteng Juchuang released the all-solid-state blind lidar RS-LiDAR-E1;11 month, and Wesai Technology officially released the pure solid-state lidar FT120 for ADAS front-loaded production vehicles.

Tudatong's Falcon semi-solid lidar claims to be the farthest lidar on the market and the most important reason for the rapid growth of Chinese companies is the fierce competition and large number of participants in China's self-driving.

Just like the Velodyne was supported by the self-driving program in the United States. Only soil can give birth to more seeds.

According to statistics, since 2021, the total financing of the global lidar market has exceeded 6 billion yuan, of which Chinese companies have received 5.8 billion. There are 21 lidar models worldwide and 16 Chinese companies.

The booming development of vehicle lidar in the Chinese market is unanticipated by Velodyne.

In 2019, Velodyne chose to shrink strategically. In order to focus on the European and American markets, it withdrew its offices in China after global layoffs, and the direct sales model returned to the "agency model" that had just entered China.

At present, Velodyne's customers in China still have self-driving companies focusing on smart trucks, such as Zhijia Technology, mainline Technology and so on. But Velodyne offered rare particularly favourable business terms in order to stabilize China's truck market, according to people familiar with the matter.

NO.3 [the rice bowl taken away by Tesla, and the uncertain future]-Why have bats evolved so long that they don't even want their eyes and use ultrasound?

Let me ask you, do more animals use their eyes or do they use ultrasonic waves?

This is a classic conversation about whether the ultimate autopilot scheme will have lidar.

Although lidar is very popular in recent years, the road ahead of lidar has not become clearer and clearer. Similar discussions have existed all the time.

At present, the basic consensus in the industry is that lidar is one of the necessary hardware to help vehicles achieve high-order intelligent driving functions.

But as Musk takes the visual route more and more thoroughly, in the past year there has been more and more talk in the industry about when lidar and high-precision maps will be abandoned.

Musk once said, "lidar is a fool's errand, relying on lidar is doomed to failure."

Tesla's reason for rejecting radar, in the final analysis, is that they are not fast enough to reduce costs and increase efficiency. On the B side of the story, Tesla bet on pure visual route technology by leaps and bounds, so fast that he brought all kinds of radar jobs in less than three years.

Many companies have begun to stand in line with pure visual routes, such as Zhijia Technology. This is a self-driving company headquartered in Silicon Valley with offices in Suzhou, Beijing and Shanghai. One of their R & D engineers said frankly, "We are currently using lidar, but we are watching the path of Boss Tesla."

Although many other car companies still have more lidar than others today, many people say that it is an expedient measure, and most of them are targeting Tesla's route, and there is a good chance that they will reduce the use of lidar in the future.

For example, when asked if lidar is the core hardware? Most car companies replied: "just do some redundancy."

Of course, there must also be a situation of "mouth disrelish and straightforward". After all, the ability of the software can not be reached, so it can only be put together by hardware. It is hard to say whether lidar can really be abandoned completely in the future.

Despite the long-term threat of pure visual route, lidar companies still have a lot of problems.

"their business volume is almost interesting," said an investor related to the auto industry. in September this year, Wo Sai announced that the monthly delivery of lidar exceeded 10,000 sets, which is still the fastest head player. "

On the profit side, he commented: "all are unstable, and the range of losses tends to expand. Wo Sai had planned to be in Science and Technology Innovation Board IPO, but then he retreated himself."

On the one hand, those lidars that are already on the bus are likely to be "damaged at any time".

Zhu Xizan, a professor at Tongji University and director of the Automotive Safety Technology Research Institute, told us: "I have participated in the purchase of lidar in some demonstration zones, and none of them dare to promise a warranty period, not even an one-year warranty."

After the damage, there is also a lack of reasonable maintenance plan.

This year, a maintenance list of Xiaopeng P5 accident has pushed lidar to the forefront of the wind and waves. The list shows that the price of a single lidar is 8916 yuan, including the cost of man-hours and support materials, and the maintenance and replacement of a lidar is close to 10, 000 yuan. All of a sudden, lidar became the "assassin" who stabbed the wallet.

Tudatong-1 lidar products are said to have a high repair rate, although the ultimate prospect is uncertain, it is certain that the lidar's short-term prospects are bright.

After all, at present, high-precision map + lidar is a scheme that can quickly produce stage results. It will be an essential part of the autopilot program until it is overtaken by the visual solution.

Yole also predicts that the automotive ADAS lidar market will grow rapidly in the next five years, with an average annual compound growth rate of 73%. By 2027, the size of the ADAS lidar market will increase from US $38 million in 2021 to US $2 billion in 2027.

This article comes from the official account of Wechat: AutoReport Automobile production Jing (ID:autoreport), written by Liang Qiumeng and edited by Yu Jie

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.