Click to follow if you like!
Q: What are the core products and positioning of Orbbec?
A: The core product is the spatial intelligent measurement device (used to measure the distance between objects and the human eye), which is an essential foundational tool for spatial intelligence, aligning with Fei-Fei Li’s mention of humanoid robots and the next frontier of AI—spatial intelligence.
Q: What are the downstream application scenarios and order status for Orbbec?
A: Application scenarios include robotic vacuum cleaners, offline payments, industrial cameras, consumer electronics, 3D printing, and humanoid robots (such as those tested by Tesla); orders are blooming in multiple areas, with potential revenue from a single segment reaching hundreds of millions to billions, currently holding orders that support high revenue growth.
Q: What are Orbbec’s performance and profit expectations?
A: Performance is expected to double each year, with this year’s revenue projected at 1 billion, next year exceeding 1.6 billion, and the year after exceeding 2.5 billion; the net profit margin is expected to reach 20% this year, aiming for 30% in the medium to long term, with an overall performance growth rate expected to exceed 90% over the next four years, and next year’s profit expected to be at least 320 million with room for upward adjustment.
Q: Where does Orbbec’s competitive advantage lie?
A: It is the global leader in the field of spatial intelligence, with a domestic market share of over 50%, and over 70% in the South Korean service robot market; competitors include Intel’s related teams that have been defeated, capturing RealSense’s market share, with scattered solution patents monopolized by Apple, forming a technical barrier.
Q: What are the growth drivers for Orbbec?
A: The penetration rate of robots in the industry is increasing; the number of cameras mounted on a single robot is increasing (from 1 to 4, with a single machine ASP reaching 8000 yuan); the company’s market share continues to rise, with AIoT (gross margin over 50%) and overseas business (gross margin over 60%) increasing their share, boosting overall profitability.
Q: What is the core logic behind recommending Hengli Hydraulic?
A: It has strong capabilities in core components (first-tier in China), with low implicit valuation for robots; it is compatible with Tesla’s mass production, participating in synchronous R&D, with strong certainty in pinpointing; the main business of construction machinery is entering an upward cycle, with clear performance growth.
Q: What progress has Hengli Hydraulic made in the robotics business?
A: Samples have been sent to Tesla for the body, including 4-bar and dexterous hand micro 4-bar, synchronously developing large water screw planetary ball screw solutions; plans for expansion are underway, with short-term layouts for 20,000 to 50,000 units next year, and 12 imported high-precision machine tools ordered, expecting to reach 20-30 production lines by the end of Q3 next year.
Q: What are the performance expectations for Hengli Hydraulic’s main business?
A: Revenue is expected to grow by 20% next year, with profit growth of 25%-30%; the excavator segment is expected to grow by over 20%, while the non-excavator segment is expected to grow by about 15%; in Q4, the production of excavator cylinders is expected to increase by 40%, with revenue expected to grow by 10%-20%, and next year’s profit expected to be 3.6 billion, with valuations at historical lows.
Q: What are the key dynamics in the field of embodied intelligence recently?
A: Fei-Fei Li’s team released the Marvel world model, promoting the switch of AI to spatial intelligence; the North American startup Generalist released a 10 billion parameter model “Zhenling”, verifying that embodied intelligence conforms to the scale law, with 7B parameters as the activation threshold.
Q: What are the core features of the “Zhenling” model?
A: The parameter count is 10B, pre-trained on 270,000 hours of real machine data (with an additional 10,000 hours added weekly), not relying on simulation or human video data; it adopts a “harmonious reasoning” architecture, integrating System 1 and System 2 to enhance response real-time; it reduces the cost of real machine data collection through the UMI collection mode.
Q: Why is 7B parameters a key threshold for embodied intelligence models?
A: Below 7B (700 million) parameters, model performance is non-linearly related to data/computational power growth, leading to potential regression; above 7B, rapid generalization and adaptation to new tasks can be achieved through minimal post-training, and since embodied intelligence involves multiple dimensions such as language, vision, and touch, the threshold is 1-2 orders of magnitude higher than traditional LLMs.
Q: What are the advantages of the UMI data collection mode?
A: Low cost (less than $500 per set), collected by a handheld claw equipped with a panoramic camera; high data quality (real environment interaction), only collecting information from the end of the hand, avoiding cross-robot body transfer issues, balancing generalization and low cost.
Q: Which companies related to embodied intelligence are recommended by the computer team?
A: On the model side: Suochen Technology (robotic cerebellum), Nengke Technology (end-side landing and command platform); on the sensor side: Suteng Hesai (leading in LiDAR, focusing on 2D to 3D laser vision integration).
Q: How is the recent market and industry rhythm for the robotics sector?
A: Since October, infrastructure and robotics stocks have corrected by over 20% (due to market sentiment factors); the industry rhythm is progressing smoothly, with Tesla’s mass production next year being a significant milestone for the industry, and the first batch of designated suppliers will be announced from late November to early December; domestic related companies have completed ITOT guidance, with ample catalysts ahead.
Q: What are the current investment focus directions in the robotics industry chain?
A: Focus on companies with core component capabilities, high market share, and high card position tracks; prioritize domestic suppliers that can adapt to mass production for leading customers (such as Tesla), have positive R&D capabilities, and reasonable valuations.
Disclaimer: This article is compiled from company announcements, news, publicly available research reports, and social media networks, and does not constitute investment advice for the industries and stocks mentioned. The market has risks, and investment should be cautious. If there are any infringements or violations of information disclosure rules in the images and texts, please contact us to delete them.