On December 2, robot-related stocks experienced widespread growth, with Juneng Co., Ltd. surging 30%, and T Starr , Buke Co., Ltd., ASDA, Gugao Co., Ltd., Huaxingyuanchuang, and Zhongdalide all following suit.

It has to be said that the robot sector is becoming increasingly crowded.

In March 2023, Samsung invested over RMB 300 million in domestic robot manufacturer Rainbow Robotics. This company's robot, HUBO, is South Korea's first bipedal walking robot.

From Boston Dynamics, Honda, and Tesla to OpenAI and Microsoft, with the advent of the generative AI era, a new competition surrounding robots has begun.

Humanoid robots are not a new concept. In 1973, Waseda University's Professor Katoh Ichiro led a team in developing the world's first life-sized humanoid robot, WABOT-1. However, half a century later, commercialization remains elusive, with performance and cost being the core obstacles.

Robots are composed of three key technical modules: motion, sensing, and artificial intelligence. For traditional robots, often only one of these technologies is required for functionality. For example, industrial robots focus primarily on motion control technology, while robotic vacuum cleaners emphasize navigation and sensing technology.

Humanoid robots, however, must be versatile in application scenarios rather than performing single tasks in specific contexts. This adds complexity, requiring higher technical integration, larger dataset modeling, and stronger language and command comprehension. Prior to this, AI data and models developed in isolation, with slow iteration speeds and persistently high costs.

The proliferation of large models has significantly altered this landscape.

From the 2017 Transformer to GPT-1, BERT, GPT-2, GPT-3, GPT-4... the number of model parameters has soared from billions to trillions in just a few years. Consequently, large models are evolving from single-modality (text, speech, vision) to multi-modal general AI. This facilitates the direct integration of speech, vision, decision-making, control, and other technologies with humanoid robots, comprehensively enhancing their capabilities.

In April 2023, AI company Levatas collaborated with Boston Dynamics to integrate ChatGPT and Google's text-to-speech technology into the Spot robot dog, enabling successful human interaction.

The rapid evolution of underlying technologies has demonstrated the potential for large-scale commercialization of humanoid robots. Major global technology companies are actively exploring and preparing for this. However, currently, humanoid robots are still some way from becoming a household item.

Firstly, the capabilities of humanoid robots currently available on the market are relatively limited and lack substitutional benefits.

For instance, household service humanoid robots cannot fully replace existing human services, and commercial guide robots can only answer simple questions, unable to address all customer inquiries.

Due to this lack of rigid substitutability, humanoid robots remain unappealing to consumer-end (C-end) users.

From the product positioning of participants, it is evident that current efforts are primarily focused on exploring B-end scenarios. For example, Tesla's first batch of mass-produced robots will likely be deployed in superfactories, while Unitree's focus is on exploring humanoid robot applications in industries such as new energy vehicles and 3C electronics through corporate collaborations.

Secondly, many foundational technologies still have shortcomings. For example, robot batteries need to support continuous operation for up to 20 hours, but most current humanoid robots can only operate continuously for less than 2 hours. Additionally, the cost, often tens of thousands of dollars, is currently prohibitive for widespread adoption, necessitating a 15%-20% annual reduction in manufacturing costs.

Humanoid robots may not be a lucrative business in the short term (3 years), but they represent an irresistible industry in the long term (30 years). From an industrial first-principles perspective, the value of humanoid robots lies in replacing high-cost human labor, a highly certain outcome.

Current technological and cost issues will diminish over time.

Cost-wise, as with computers, smartphones, and electric vehicles, any new technology or product will inevitably transition from high to low prices with continued commercialization. Initially, the cost of humanoid robots like Honda's ASIMO and Boston Dynamics' Atlas was as high as USD 3 million and USD 1.9 million, respectively, but Tesla has reduced this to USD 20,000, with further reductions anticipated.

Similarly, technology will continue to iterate and improve. Due to immature lithium-ion battery technology, the first-generation humanoid robot Atlas by Boston Dynamics in 2013 required a power cable, but the second-generation Atlas in 2016 utilized independent lithium-ion batteries.

Elon Musk predicts a long-term demand for 10 billion humanoid robots. Even if only one-tenth of this demand is met, the industry's potential is enormous.

With the largest population and manufacturing sector, China is already the world's largest robot consumer market, with robust demand anticipated for both B-end and C-end applications. Moreover, China's technological gap in robot hardware and software is not significant. These factors fundamentally position China to cultivate world-class robot companies.

Currently, China's leading humanoid robot enterprises are Unitree and CloudMinds, founded in 2012 and 2015, respectively. Realistically, however, domestic giants like Xiaomi and ByteDance are more likely to emerge as significant players in the humanoid robot industry due to their advantages in talent, capital, market access, and branding. Although Unitree and other early starters have a head start, they have not accumulated a significant lead, with Unitree's humanoid robot revenue accounting for only a single-digit percentage of total revenue.

Upstream visibility and profitability are significantly stronger. Analyzing the material costs of robots reveals that reducers, servos, and controllers account for 35%, 20%, and 15% of industrial robot costs, respectively, totaling 70%. Considering humanoid robots have more joints and degrees of freedom, these components may account for an even higher proportion.

The reducer market is crowded with players like Dazhu Transmission, Leadshine, Tongchuan Technology, Zhongdalide, and Guomao Co., Ltd., but only one dominates: Leadshine. The company has established a virtuous cycle of "R&D - expansion - profitability - further R&D - further expansion." By 2022, Leadshine's annual production capacity reached 400,000 units, with potential expansion to 590,000 units in 2023. In comparison, Tongchuan Technology, Dazhu Transmission, and Guomao Co., Ltd.'s annual capacities in 2021 were only 100,000, 60,000, and 30,000 units, respectively. As a manufacturing enterprise, Leadshine achieves a net profit margin of over 30%, which is remarkable.

The servo motor market is more stratified, with high-end capacity largely controlled by foreign companies like Mitsubishi, Yaskawa, Fanuc, and Siemens. Domestic companies like Inovance, Jiangte Motor, Jiangsu Leili, Leadshine, and Hozon Tech focus on the mid-to-low end. Among them, Inovance is the clear leader, with a 2022 domestic market share of 21.5%, up 5 percentage points from 2021. Anticipated domestic substitution and increased robot production bode well for Inovance's future prospects.

The domestic controller market is fragmented, with companies like Canopen, Wanxun AutoControl, Gootech, Inovance, and Haide Control among the players. However, they have yet to form an effective market competitive force, with a domestic production rate of less than 20%. It remains to be seen whether any of these companies will emerge as significant players.

In the long run, humanoid robots are an inevitable future trend. Short-term speculation is meaningless; what is needed is long-term tracking and attention to identify key enterprises.

Disclaimer

This article contains content related to listed companies based on the author's personal analysis and judgment of information publicly disclosed by these companies in accordance with legal requirements (including but not limited to interim announcements, periodic reports, and official interaction platforms). The information or opinions contained herein do not constitute any investment or other business advice. Market Value Observation assumes no responsibility for any actions taken based on this article.

——END——