Musk’s humanoid robot, VCs have long been eyeing

“Volume King” Musk really came with his humanoid robot Optimus (Optimus Prime). In addition to the National Day, my circle of friends is Optimus swiping the screen.

Knock on point: Optimus, which could target prices below $20,000, is cheaper than Tesla cars, uses the same computing hardware as Tesla’s self-driving cars, and will eventually produce millions of units.

Musk wants to disrupt the robotics industry as an outsider, just as he has disrupted the auto industry before.

Before Optimus was unveiled, I chatted with people in the industry, and the consensus was that robots will undoubtedly be the big track for the next few decades, and humanoids are still second, but the “deep essence” pioneered by Tesla “The style may affect an era.

And, most notably, there are already many investors looking for “Wei Xiaoli” in the robotics industry. In particular, with the advantages of a stable domestic supply chain and controllable delivery capabilities, domestic robots can fully seize the opportunity of rapid changes in the current industry and successfully achieve corner overtaking.

Screenshot of the source conference

Overtaking on a curve but still need help

After seeing Tesla’s humanoid robot, many people said that it was very similar to when Apple first released the iPhone.

What’s the meaning? After the various effects are displayed, the Optimus that attracts the most attention is actually a “prototype”, and the mass production time is not yet clear. In Musk’s words, the technology will definitely be further upgraded, and conversations and conversations can also be conducted, but Just don’t know what the ultimate goal is.

And when it was displayed, an Optimus prototype did not carry a shell, but directly “roughly” showed its internal structure, and even now it needs three people to assist, and it was also “the first time without a tether. walk”.

To be on the safe side, Musk didn’t even let the Optimus prototype make more action demonstrations. As for further practical functions, he used demo videos to show them, such as moving cardboard boxes in a car factory, watering plants, moving metal rods, etc. .

Another relatively perfect prototype of the Optimus is also, although it looks more refined and fully assembled, it is also a later production prototype, but the function is not fully functional. In Musk’s words, “it’s not quite ready to walk, but I think it will be in a few weeks.”

Of course, Musk’s speed is already fast, because a lot of experience in car building has been transferred to robots. According to the introduction, Optimus uses the same computing hardware as Tesla’s self-driving cars, such as Optimus “brain” running on Tesla On the SoC (system-on-chip), the Autopilot system for navigation is similar to that on electric vehicles, in addition to battery packs, cooling systems, etc.

Musk also said that Tesla has successfully made a four-wheeled robot-car. This success is indeed obvious to all, and it is a good idea to reuse the increasingly perfect technologies in auto-driving in the field of robots that also need these capabilities. Very reasonable route.

And he also wrote in his contribution to “China Netcom”, “When one day we solve the problem of autonomous driving of cars (that is, AI problems in the real world), we can extend AI technology to humanoid robots. It will have broader application prospects than cars.”

As for why Musk is so aggressive? In April, Musk said in a public interview that Optimus was Tesla’s “second curve” for growth because the self-driving business application fell short of his expectations.

So when the cost of the Xiaomi humanoid robot still needs 600,000 to 700,000, Musk has put down his rhetoric: the target price of Optimus (Optimus) may be lower than 20,000 US dollars, and more than one million units will be mass-produced.

Screenshot of the source conference

Humanoid robot track still has a long way to go

But I still want to pour a pot of cold water when I am swiping the screen in the circle of friends. I can come up with these robot performances without even going to the robot conference, and most people in the industry and even more people have realized that , the performance of humanoid robots has completely lost its freshness.

Before Tesla, Boston Dynamics, which was once specialized in the former, changed hands several times because of the dismal sales of robots, and even after the commercialization of the detection robot dog Spot was confirmed, the business situation eased? And after the acquisition, Hyundai Motor even stated that it only hopes to use the technical capabilities of Boston Dynamics robots to help its related projects succeed.

Even due to the lack of commercial value, Honda gradually stopped the research and development of humanoid robots as early as 2018. Softbank also announced in June this year that the humanoid robot Pepper was discontinued, and began to cut robot development teams in France and Japan.

Therefore, although “outsiders” come in, it still cannot be implemented on a large scale in a short period of time. Humanoid robots still have many bottlenecks such as long research and development cycle, high cost investment, difficult coordination of refined technologies, and difficulty in information security and user security.

Turning your attention back to China, more than a month ago, Xiaomi’s first full-scale humanoid bionic robot CyberOne, known by the internal stage name “Tie Da”, was officially unveiled. In addition to Xiaomi, CloudGinger recently released its latest version of its flagship humanoid service robot, Cloud Ginger 2.0.

But Lei Jun only said at that time that CyberOne is learning new skills every day. In the field of intelligent robots, humanoid bionic robots have the highest degree of technical integration and the most difficulty, and Xiaomi is still in the first stage of its infancy.

Of course, the long way to go does not mean that the humanoid robot track has no merits. As mentioned at the beginning, Song Yujie, a hard technology investor, also shared with me that the prototype of Tesla Bot has not been released, and there are many robots in the secondary market. The company’s market value has doubled. In the primary market, everyone is looking for the next batch of “Wei Xiaoli” that is the benchmark for Tesla Bot.

Let me try to explain, this may be divided into two aspects:

On the one hand, the upstream and downstream of the industrial chain involved in humanoid robots have obtained substantial benefits, such as rising stock prices, rising market value, etc., and the impact is immediate. For example, Eston, the leader of domestic industrial robots, said in September that it is being frequently investigated.

The research report of Guolian Securities also shows that humanoid robots may benefit from reducers, motors, structural parts, semiconductor chips and devices, sensors, panel glass and materials, which will have more investment opportunities.

On the other hand, for start-ups, in Song Yujie’s words, there is still a very big opportunity to go straight to the end product definition + bottom-level innovation + Quanzhen original research core components.

Gengxin Capital China also gave opinions in its own article. It has always believed that AI and robotics will bring a huge opportunity for economic growth to the world; in the second half of the information technology revolution, the development of AI and robotics will be the answer to this The most problematic variable.

Therefore, on the whole, the humanoid robot with a long way to go is actually a microcosm of the robot industry.

More certain opportunities are now worth investing in

The question is, the popularity of the robot track is not a day or two. Where is the focus of investors or the popularity of the industry?

I think there is a poem that is very appropriate: If you want to learn poetry, your work is beyond poetry.

I have communicated with many investors and found that in fact, many institutions now look at the robot track, and each agency has different preferences, but in the robot track, the underlying core technologies are the same, and these alone are already a huge investment opportunity. .

And this is also in line with the development of the national industry. At the end of last year, my country released the “14th Five-Year Plan for the Development of Robot Industry”, which proposed that by 2025, my country will become the source of global robot technology innovation, high-end manufacturing agglomeration and a new highland for integrated applications. The first thing mentioned here is to promote a number of breakthroughs in the core technology of robots and high-end products, and the performance and reliability of key components will reach the international level of similar products.

In addition, in fact, the direction of investment and financing is also very telling. In the “China Robot Industry Development Report (2022)” released recently, industrial robots are still the category with the highest proportion of financing, not the service robots where humanoid robots are located.

In fact, this is not difficult to understand. As Gengxin Capital China said, for a long time in history, only a very small number of industries (such as automobile manufacturing and warehousing and logistics) have applied robots, which are characterized by high stability, long life, stable It’s good, can’t make mistakes, and it’s risky to choose a new solution. That is, in the past five years, the robot scene has been fully rolled out, and all walks of life and people have begun to use robots.

Moreover, with the outbreak of domestic demand for semiconductors, automobiles and other emerging industries, it has also driven new demand for industrial robots. According to the statistics of MIR Rui Industry, in the first half of this year, the shipments of industrial robots used in the lithium battery and semiconductor industries have increased by 20% year-on-year. 37%.

Previously, Wang Tianmiao, honorary director of Beihang Robotics Research Institute and president of Zhongguancun Zhiyou Research Institute, also mentioned in an interview with China Investment.com that China is making traditional manufacturing, including automobiles, mobile phones and machine processing, and is moving towards a new industrial direction. Development, including new energy and new energy vehicles, power batteries, new infrastructure, medical masks and masks, milk and dairy products, hot pot dishes, etc., suddenly require the assistance of robots.

The reality is that these are basically the needs of small and medium-sized enterprises, and the four major foreign families are often slow to respond. Considering that the scale is not large for the time being, the cycle of research and decision-making, product development and implementation is often long, and new Chinese innovative companies respond quickly, often from Non-standard projects start, enter the market quickly, and then gradually standardize, modularize and scale.

In addition, from the perspective of future development trends, China’s manufacturing industry is very strong. If it wants to move towards the global cycle, especially the Belt and Road, under the circumstance that the level of local workers is low or the labor cost is constantly increasing, China wants to go out, and also It has to go out with design, manufacturing automation and robotics.

In this regard, IFR statistics also show that in 2021, China’s industrial robot installation will reach 243,000 units, almost equal to the sum of the rest of the world. And by next year, the scale of my country’s industrial robot market will exceed 10 billion US dollars.

The premise of all this is to solve the reality that the underlying operating system and chips, core components, and system design software of robots rely on foreign countries for a long time, so as to better play the advantages of domestic robot enterprises. And where the shortcomings are, there are growth opportunities, and where the market is, there will be investment opportunities.

Of course, the good news is that data from the Ministry of Industry and Information Technology shows that by the end of 2021, there will be 101 companies specializing in the field of robotics, specializing in new and small giants, covering all kinds of companies such as the core components of the whole machine and system integration.

And related data show that there will be 200 investment and financing events in the robot track in 2020, rising to 288 in 2021, and 170 investment and financing events in the industry in the first nine months of 2022.

This has also spawned more and more robot unicorns, large-scale financing, and even many front-line institutions after each.

For example, Jieka Robot, which completed its Series D financing of nearly 1 billion yuan in July, this round of financing was jointly led by Temasek, Truelight Capital, SoftBank Vision Fund Phase II, and Saudi Aramco’s diversified venture capital fund, Prosperity7 Ventures. Investment, strategic investment institutions follow investment.

After industrial robots, because of the broad industry, service robots and special robots are also constantly gaining the favor of capital.

For example, Siling Robot, which became a unicorn in only 3 years and 2 months, received investment from Hillhouse Ventures and Linear Capital less than a month after its establishment, and received 5 financings from 2019 to 2021. Moreover, there are Hillhouse and Linear Capital behind the multiple rounds of financing.

Therefore, I also hope that Tesla’s humanoid robot will bring the “catfish effect” to the domestic robot industry and generate more business and investment opportunities. The increase in orders for parts and components also means the entry of more talents and funds, and it means that the turning point of the industry is approaching.

In the end, I hope that over time, whether it is a humanoid robot or other robots, it can bring long-term business value beyond the imagination. After all, no matter what the shape, the machine does bring better efficiency, and More free our hands. (Text / Source of Zhang Lijuan / China Investment Network)