Advanced Robotics Components

Yes! The first phase of our technical robot benchmark is complete: 3 humanoid robots, 350+ technical data points, and conclusions for operators and investors that differ from the market narrative. As humanoid robots scale, industrial users, operators, and investors need to deeply understand their technology. However, public technical benchmarks for humanoid robots are still rare. So we started one (for our investments): we are building an ongoing benchmark series for industrial humanoid robots — strictly based on technical analysis, not demos, headlines, or AI claims. Our first desk research benchmarking phase compares the best-selling humanoid robots for industrial use: ➡️ Unitree H1-2 ➡️ AgiBot A2 Ultra ➡️ UBTECH Walker S2 We compare them across six categories: 1️⃣ Platform architecture 2️⃣ Actuation, joint performance & locomotion 3️⃣ Manipulation, hands & payload 4️⃣ Perception & sensor stack 5️⃣ Compute, software stack & AI layer 6️⃣ Power, charging & runtime High-level summary of the findings of this phase (350+ data points, comment BENCHMARK if you want a copy.): ✅ UBTECH Walker S2 currently presents the strongest overall package for industrial use: multi-shift architecture with autonomous battery swap in <3 minutes, the strongest heavy manipulation package with 15 kg payload, a 0–1.8 m workspace, 7.5 kg single-hand grasp, and 1 kg finger grasp, etc. etc. ✅ AgiBot A2 Ultra shows the most advanced SW / AI platform for integration into a broader agentic physical AI setup: the strongest developer-to-deployment stack, including AimRT, AimDK, HTTP JSON RPC, ROS2 Topic interfaces, ROS2 Humble with FastDDS, etc. etc. ✅ Unitree H1-2 is currently more a general robotics platform than a focused industrial deployment package. However, it comes with with the strongest hand-level tactile and precision capability via the dexterous hand stack and the strongest lower-body raw actuation reserve, e.g. knee torque around 360 Nm. Current bottom line: ➡️ Best suited for industrial deployment today: UBTECH Walker S2 ➡️ Most modern overall platform: AgiBot A2 Ultra We turned the first benchmark results into a technical fact sheet: • 15+ pages • 6 technical categories • 350+ technical data points • directly comparable specs We created the fact sheet for our own use, but we are happy to share it if anyone is interested (we need to be connected). Comment BENCHMARK if you want a copy. This is only the start: we will update and expand the benchmark, add more humanoid robots as they become industrially relevant, and over time move toward teardown-based and physical comparative analysis. // No investment advice and no guarantee as to the completeness or correctness of the provided data. //

Robotic Actuation: Minding the Gap(s) In high-performance robotics, "Minding the Gap" isn't just a London Tube slogan—it’s the difference between a standard actuator and a world-class one. While these components drive over half of a robot’s cost, the leap from 'good' to 'great' isn't a single breakthrough; it’s a disciplined refinement of the entire system. To illustrate, consider the impact of optimizing these three critical gaps: ⚡ The Electromagnetic Gap  Shrinking the air gap between the rotor and stator minimizes magnetic reluctance, maximizing torque density and efficiency. To go from good to great, you must achieve extreme structural stiffness and precision to prevent catastrophic stator strikes during high-load deflections. 🔥 The Thermal Gap Air is a thermal insulator. In the high-transient world of robotics, internal air pockets lead to rapid winding burnout. Transitioning to a world-class design requires maximizing "copper fill factor" and using thermally conductive potting to create a low-resistance path for heat to escape. ⚙️ The Mechanical Gap Gearbox backlash—the "mechanical gap"—creates control deadbands and damaging "hammering" effects during impacts. A great actuator minimizes this gap to maintain the proprioceptive transparency needed for a robot to "feel" the world, requiring absolute mastery of manufacturing tolerances and metrology. 📍The Bottom Line  Modern actuation isn't magic; it is the disciplined optimization of the fundamental tensions between physics, manufacturing precision, and cost. We can design for tight air gaps on a CAD screen, but achieving them consistently on the assembly line is another story. For those building at scale, how are you balancing need for precision with the realities of high-volume manufacturing? #Robotics #Actuation #LondonTech #Manufacturing

This exploded view breaks down a complete Industrial Robotic Arm System — from base to gripper — revealing the precision engineering that powers modern automation. 🔍 Key Highlights: • Base & J1 Axis – Provides stable 360° rotation (Z-axis), forming the robot’s foundation • Servo Motors + Gear Reducers – Deliver high torque with precise motion control • Harmonic Drives – The real game-changer for zero-backlash, high-accuracy positioning • Arm Linkages (A2, A3) – Enable reach and flexibility across multiple axes • Wrist Mechanism (Roll, Pitch, Yaw) – Allows complex orientation for intricate tasks • End Effector (Gripper) – Where the action happens — handling, picking, assembling 💡 What makes this fascinating is how mechanical design + control systems + electronics come together to create ultra-precise, repeatable motion — the backbone of smart manufacturing. In today’s world of Industry 4.0, robots like these are not just machines — they are productivity multipliers driving efficiency, quality, and scalability. 👉 Whether you're into robotics, PLCs, or automation engineering — understanding the internal architecture gives you a real edge. #IndustrialAutomation #Robotics #Industry40 #AutomationEngineering #SmartManufacturing #ServoMotor #HarmonicDrive #MechanicalDesign #EngineeringLife #PLC #FutureOfWork #ManufacturingInnovation #TechExplained #RoboticsEngineering #Automation

NASA SBIR Ignite Funding Opportunity: #Robotics Subtopic I finally went and downloaded the topic pre-release, and I am thrilled to share that there is an NASA - National Aeronautics and Space Administration #SBIR ignite robotics subtopic this year! https://lnkd.in/gq_rQveg Subtopic I04.01: Modular, scalable robotic subcomponents to unlock scalable robotic manufacturing & assembly in remote, challenging environments One of the key challenges in robotics today is the lack of standard, modular subcomponents that enable robotics to scale (actuators, motors, tools, end-effectors, beams/tubes for arms, wheels, etc). To reduce the cost of robotics for manufacturing and assembly, NASA needs basic robotic components with standardized mechanical and/or electrical interfaces that are qualified for use on orbit as well as lunar and planetary environments. The components should have the following characteristics: • Reconfigurable with non-proprietary, standardized interfaces • Allows the use of custom components designed by the end user • Designed for use in remote or challenging environments • Optimized for cost-effective mass production • Ability to be quickly scaled NASA is especially interested in solutions that balance readiness for eventual space deployment with near-term manufacturability and commercial viability. While full qualification is not required at this stage, a plan for space environment compatibility and scalability will strengthen the proposal (such as exposure to dust, vacuum (lubricants especially), radiation, UV, thermal, gravity, atomic oxygen, etc). Additionally, component approaches that demonstrate a clear path towards a complete robotic system-level solution are preferred. Considerations: • Solutions designed to be robotically assembled are encouraged. • Solutions at a scale appropriate for small-sat or orbital/surface asset aggregation applications are of particular interest. • Robot architecture is non-specific (inchworm/climbing robots, rover-based systems, free-fliers or other) • Adapting and qualifying existing robotic elements and systems for NASA applications is encouraged. Existing hardware could be upgraded to take the key elements of the design and add the components to survive challenging environments. • Examples of desired improvements in capabilities: • protection from dust and/or resilience to dust getting into moving components including dust repellant technologies and coatings • solutions capable of a range of torques • optimized lubricants for wide temperature ranges and minimal mass loss and outgassing in vacuum • solutions that use lower cost metals in order to reduce overall cost • tough, conductive thermal coatings and treatments that can resist erosion or surface damage

THIS JOINT COULD PROPEL ROBOTICS FORWARD – AND NO ONE IS USING IT❓ Three years ago, Japanese researchers Kazuki Abe, Kenjiro Tadakuma, and Riichiro Tadakuma introduced the ABENICS 3D joint. An active ball joint with three rotational degrees of freedom, achieved through a precise combination of bevel and specially shaped gears. It moves like a human shoulder or hip – no slip, no sensors, high torque, and impressive positional accuracy. As an engineer, I see this as a dream component for humanoid robots, especially in the shoulders, where space is limited and mobility is critical. And yet: no production model, no prototype at trade shows, no visible implementation in industry. Why does something like this remain in the lab? Is it a lack of courage to replace existing solutions? Manufacturing challenges? Or have we as an industry become so comfortable with “good enough” that we stop implementing real breakthroughs? My view: Any technology that delivers precision, strength, and compactness in a single unit deserves to make the leap into real-world applications – otherwise, we lose years of progress. What do you think: Is the problem the technology, the market, or us engineers? Follow me if you want to see the technologies that deserve to break out of the lab – and share your perspective in the comments. Best Regards #CobotUli ULMO Consulting & Marketing #Robotics #Innovation #Engineering I post at 8:00, 11:30, and 17:00 (Berlin time) 3x daily, 100% real robotics. No fluff. No filters. No fakes. PS: LinkedIn hides most posts. 👉 Join my group or miss out: https://lnkd.in/e9skpAF

TECHNOLOGY BEHIND ATHLETIC ROBOT WITH KUNG-FU SKILLS. 1. Athletic kung-fu robots combine advanced AI with biomechanical engineering to mimic martial arts movements. 2. Their bodies are made from lightweight, flexible materials to replicate human muscle dynamics. 3. Servo motors and hydraulic actuators provide precise joint control for high-speed kicks and punches. 4. Real-time motion tracking systems guide balance, posture, and reactive responses. 5. Embedded gyroscopes and accelerometers help maintain stability during flips, spins, and rapid stances. 6. Machine learning trains the robot using thousands of kung-fu move datasets. 7. Computer vision allows the robot to analyze opponents’ positions and adjust attacks dynamically. 8. Their limbs are powered with multi-axis rotation systems for fluid and complex motion. 9. Force sensors ensure controlled impact during strikes and block maneuvers. 10. The robot’s software emulates martial arts strategy, timing, and rhythm. 11. Its core system adjusts center of gravity for realistic balance and posture shifts. 12. Voice recognition enables command-driven sparring routines and performances. 13. Dynamic torque control prevents joint damage during intense movements. 14. Reinforcement learning improves the robot’s kung-fu performance over time. 15. Articulated feet provide grip and shock absorption on different surfaces. 16. The robot can perform acrobatic flips and evasive rolls using gyrodynamics. 17. Motion capture of human masters is used to replicate authentic kung-fu forms. 18. Some robots are designed to spar with humans using real-time response systems. 19. These robots showcase potential in education, performance art, and defense training. 20. Kung-fu robots demonstrate how AI and robotics can master traditionally human physical skills with agility and grace.

Here's how to build autonomous ground robots for navigating challenging environments. 🔩 Reliable hardware High quality components must be sourced, and arranged in position with tight tolerances. A lot of trouble can come from just one bad cable or one suspension that's slightly misaligned. Design, assembly, quality assurance, and testing are necessary before any downstream development begins. ⚙️ Control Your motors and suspension are in place. Now what? At the low level, motor controllers must be tuned and responsive, to ensure wheel speeds enable precise robot kinematics. All sources of latency must be sought out and eliminated. At the high level, the robot must be able to follow a target path without overshooting, oscillating, or crashing into things. 👁️ Perception Cameras, LiDARs, neural networks, and algorithms must work together to create a map of the surroundings. Sensors must be intrinsically and extrinsically calibrated, and synchronized with the control stack. Neural networks require a support pipeline of labeling, dataset curation, training, edge export, and inference. Algorithms bridge it all together, mapping areas to travel to, and areas to avoid The robot businesses that solve these three problems survive. The ones that don't will fail. What has been the biggest challenge for robots in your field? #Robotics #Autonomy #Innovation #FieldRobotics #AI

⚙️🦿🦾 Actuators Actuators are the component of the robot that enable motion. There’s a wide variety of actuators to choose from, depending on energy sources, control inputs, motion types, and output needs. I have had the opportunity to work with several different kinds of actuators, from simple hobby servos to force-controlled actuators. My experience: 🤖🦿Minimal: Hobby Servos Hobby servos are probably the most accessible actuators. They are well known for being precise and easy to control, while also being relatively inexpensive. They are typically controlled by a PWM signal from a microcontroller, which maps to a specific angular value. On Minimal, these actuators were driven by a Teensy 4.1, which received desired servo positions from the Raspberry Pi running the RL trained control policy. Low-level control: The servo’s internal PID loop makes control straightforward, you only need to send a position target. However, you have very limited bandwidth and no direct torque control, which limits their use in dynamic locomotion tasks. 🤖🦿BLDC Motors with moteus Controllers For higher performance robots, I transitioned to BLDC motors controlled by moteus. These use field-oriented control (FOC) with integrated encoders and current sensors. The actuators had a 1:10 reduction planetary gearbox designed by Aniket M. I used the moteus drivers that can run position, velocity, impedance or torque control internally at 30 kHz. Low-level control: Position/impedance mode leverages the fast internal loop, giving high-bandwidth compliant behavior (critical for legged locomotion).

In the robotic palm, both the coreless gear motor and the magnet play crucial roles. Here is an introduction to their applications: • Application of coreless gear motors: The coreless gear motor is a core driving component of the robotic palm. It features high torque, high responsiveness, high reliability, high efficiency, and low temperature rise. For example, the coreless motors from Leadshine have achieved higher torque and response speed through optimized electromagnetic schemes and winding technologies. They also adopt high-reliability components and have a long brushless service life, enabling the robotic palm to be more "dexterous" and "powerful". Additionally, they can reduce the burden on the battery and improve endurance. Moreover, the coreless motors from Dingzhi Technology, with characteristics of high efficiency, high rotation speed, and small size, perfectly meet the requirements of "small size and large torque" for the finger joints of humanoid robots. • Application of magnets: Magnets are an important part of the coreless motor, usually located on the rotor, mostly in arc or tile structures attached to the rotor. They are generally made of rare-earth permanent magnet materials such as neodymium-iron-boron and samarium-cobalt, which have high magnetization intensity and good magnetic properties. The magnetic field generated by the magnets interacts with the magnetic field generated by the current in the stator windings, thereby generating torque to drive the motor to rotate. In the robotic palm, the magnets adapted to the coreless motor need to be precision-machined to be embedded in limited space in an ultra-thin and miniaturized form, while ensuring uniform and stable magnetic force. Only in this way can they cooperate with the motor to achieve micron-level precise motion control.

Figure, a leading AI-robotics company, raises $1.5B at $39.5B valuation and aims to build 100k AI-powered robots. Here’s why this matters: → Figure is developing a new AI system called Helix. It allows humanoid robots to perform complex tasks through voice commands. → Helix can handle unfamiliar objects without needing specific training for each one. This breakthrough could revolutionize household chores. → The system combines two powerful AI components. The first is a 7-billion-parameter multimodal language model. It processes speech and visual information at 7-9 Hz. This acts as the robot's brain. The second is an 80-million-parameter AI. It translates the language model's instructions into precise movements at 200 Hz. → Helix can control 35 degrees of freedom. This means it can manage everything from finger movements to head and torso control. → Figure has shown robots responding to voice commands and accurately grasping objects. One demonstration featured two robots placing food items into a refrigerator. They did this without prior training on those specific items. Training with limited data is a huge leap forward in progress in AI robotics. The system needed only 500 hours of training data. This is far less than what other projects require. The robots run on embedded GPUs, making commercial applications more feasible. CEO Brett Adcock sees Helix as key for scaling robots in homes. Unlike traditional robots, Helix adapts to new tasks without reprogramming. Its real-world performance still needs testing. Here’s their latest video showing Helix in action with two robots learning on the fly to put items away in a kitchen.