AI Second Half: A Comprehensive Analysis of the US Humanoid Robot Industry Chain White haired stock god Serenity @ aleabitoreddit says: Robotics is next In the past two years, the market has mainly traded AI brains: GPUs, CPUs, storage, data centers, big models, and inference computing power. But after the brain is built, AI cannot just stay on the screen forever, writing code, drawing, and chatting. The big model is the brain of AI, and the robot is the body of AI. The brain is responsible for thinking, and the body is responsible for action. Only when AI has a body can it truly enter factories, warehouses, hospitals, and homes to move, sort, assemble, care for, clean, and ultimately create value in the real world. Biteye has systematically sorted out the core targets of US stock robots and supply chains based on the components of humanoid robots perception system This part is responsible for allowing robots to see the world. It converts the real world into digital information through sensors such as vision, depth, and touch, allowing robots to recognize objects, understand the environment, and perform spatial positioning, which is the first step for robots to understand the external world. CGNX Cognex |~$8B+| Industrial Machine Vision Leader, Factory Automatic Inspection and Recognition System KEYS Keysight |~$30B+| Testing and measurement system, robot electronic system verification infrastructure TDY Teledyne Technologies |~$25B+| High end industrial camera+LiDAR+sensor system ON onsemi |~$25B+| Core supplier of autonomous driving and robot image sensors NOVT Novanta | $5.61B | Making robotic manpower/torque sensors is important in humanoid robots and industrial robots. AMBA Ambarella |~$4B+| AI vision chip for robot/drone/security recognition VPG Vishay Precision Group |~$1.7B | Precision Mechanics Sensor Company, strain gauges and force sensors can be used for humanoid robot dexterous hands, joint force control, and high-precision grasping, and is one of the elastic targets of Optimus supply chain. VLN Valens Semiconductor |~$200M+| is a high-speed signal transmission chip company that provides low latency visual data transmission solutions that can be used for multi camera robots, autonomous driving, and edge computing systems. execution system The execution system is equivalent to the muscles and joints of the robot, responsible for converting the instructions issued by the computing system into real actions. Whether it's walking, turning, grabbing, transporting, or maintaining balance, it all relies on motors, reducers, hydraulic/pneumatic systems, and precision transmission components to complete. PH Parker Hannifin |~$120B+| Motion control+hydraulic system, robot core drive and actuator IR Ingersoll Rand |~$32B+| Industrial Compressor+Automated Pneumatic System FTV Fortive |~$18B+| Industrial Precision Tools+Automation Equipment System RRX Regal Rexnord |~$14B+| Motor+Transmission System, Industrial Robot Power Core Calculation and Control Computing and control are equivalent to the "brain+cerebellum" of robots, responsible for processing perceptual data, understanding the environment, planning paths, and transforming AI decisions into stable actions. It determines whether the robot responds quickly, moves accurately, and can operate autonomously in complex environments. NVDA Nvidia |~$3.0T+| AI GPU+Jetson edge computing platform, the absolute core of robot computing power MSFT Microsoft |~$3.2T+| Azure+Copilot+AI infrastructure, providing robot cloud brain and development platform GOOGL Alphabet | ~$2.0T+ | DeepMind + Waymo， Layout of autonomous driving and robot AI systems AMD |~$880B+| GPU/CPU replacement solution, providing edge and training computing power for robots energy system The energy system determines how long robots can operate and how strong their power can be, including battery technology, power management, and charging systems, which are key limiting conditions for the large-scale implementation of robots. ADI Analog Devices | $203.57B | Analog chip, power management, sensing, and industrial control giant, the underlying hardware core of robots. HON Honeywell | $144.11B | Power Management+Industrial Energy Systems MPWR Monolithic Power Systems | $76.83B | Leading power management chip, widely used in robotics, AI servers, and automotive electronics. VICR Vicor | $14.40B | A high-efficiency power module that converts battery voltage into electricity for different components, suitable for robot energy systems. AMPX Amprius Technologies | $2.43B | Silicon anode high-energy density lithium battery, has entered the drone and robot markets BLNK Blink Charging | $85.63M | EV Charging Infrastructure Company, Exploring Charging Scenarios for Robots and Logistics Equipment basic software The basic software is the true "intelligent base" of humanoid robots, which determines that robots are not simply repeating actions, but can understand tasks, learn environments, and continuously optimize behavior. In the training of humanoid robots, the importance of software is even higher than that of a single hardware. Robots need to first learn to walk, grasp, avoid obstacles, and perform tasks in a simulation environment, and then continuously modify the model through real-world data. MSFT Microsoft |~$3.2T+| Azure AI+Copilot ecosystem, robot cloud brain and enterprise AI development platform. NVDA Nvidia |~$3.0T+| Isaac Sim+Jetson+GPU computing platform, robot simulation training and AI inference infrastructure. TSLA Tesla |~$1.3T+| FSD+Optimus data closed-loop, representing end-to-end robot strategy model. humanoid robot In the field of humanoid robots, the goal of the ontology company is to enable robots to complete universal tasks in complex environments like humans; In the field of industrial and collaborative robots, greater emphasis is placed on stability, efficiency, and large-scale deployment. Therefore, this layer is usually the part of the robot industry chain that is closest to end demand and easiest to form brand and valuation narratives. TSLA Tesla |~$1.3T+| Optimus humanoid robot+FSD autonomous driving unified system, representing the integrated route of "AI brain+robot body+data closed-loop". HON Honeywell |~$140B+| Industrial automation, aviation automation, and logistics automation solution provider, with more focus on robot system integration and scenario implementation. ABB |~$100B+| Global leader in industrial robots and automation, with products covering robotic arms, control systems, and production line automation, is a representative of mature industrial robots. TER Teradyne |~$24B+| is an industrial collaborative robot platform company that utilizes Universal Robots and MiR to layout collaborative robots and autonomous mobile robots. CCXI Churchill Capital Corp XI/Agility Robotics | Proposed merger valuation of approximately $2.5B | Agility is an American humanoid robot company with its core product Digit focused on logistics, warehousing, and manufacturing scenarios. At present, Agility Robotics is going public through a merger with SPAC company Churchill Capital Corp XI. After the transaction is completed, it is expected to be renamed Agility and use the new ticket AGLT. In conclusion The investment logic of the robot industry chain is actually very similar to the semiconductor cycle. In the early stages, it is difficult to determine who will become the ultimate winner, but every layer of the industry chain will have a phased opportunity: perception, execution, energy, control, and ontology, all of which will be re priced as industry attention increases. Today, Ubiquitous's super bionic robot flooded the screen, which is a very obvious signal: robots are starting to enter the public eye from laboratories and press conferences. According to public reports, Ubiquitous announced at a press conference on June 30th that the "Ubiquitous U1" series has exceeded 10000 omnichannel orders and plans to deliver them this year. Of course, this does not mean that robots will be fully commercialized immediately. In the short term, it is more like a turning point from "concept display" to "real orders". What is truly worth paying attention to is whether orders can be sustained, costs can be reduced, and the supply chain can keep up. For ordinary investors, instead of betting on who the "Nvidia of the humanoid robot world" is now, it is better to study robots as a complete industry chain: perception, execution, energy, ontology, each layer selects core companies, and disperses the uncertainty of a single target through a combination. The statement 'Robotics is next' is most likely correct. But the pace will not be achieved overnight. The real opportunity may be hidden in the core supply chain that every robot cannot bypass.