What is true on Actuation?

How do you teach a humanoid robot to walk without programming every move? Reinforcement learning step by step 1. Build a simulation (MuJoCo or NVIDIA Isaac Lab) with joints motors sensors and a physics model. 2. Define the task: walk forward while keeping balance. 3. Write a reward function points for moving toward the target staying upright and matching the desired speed. Penalties for falls, wasted energy and unstable movements. 4. A neural network, the policy, reads the robot's state (joint positions, joint velocities, body orientation) and outputs target positions or torques for each joint. 5. An algorithm like PPO updates the policy over many episodes and millions of simulated time steps. 6. Close the sim-to-real gap with domain randomization, then test carefully in the real world.

You're not ready for this. One of the most human-like humanoid robots has just entered production. Tesla says Optimus production has started on its high volume manufacturing lines in Fremont with the first production quality units coming off the line. Tesla says the current process is suitable for a commercial product, although the hardware and manufacturing methods will continue to evolve. The company has also outlined plans for a Fremont line capable of producing up to 1 million robots per year. That is a target capacity, not its current output. How quickly can Tesla move from early production to manufacturing humanoid robots at scale? And what production numbers will we see over the next 12 months? Source @ SawyerMerritt

@tomwookieford - we don’t show things that we can’t do in production. The cold gas system can produce over 80kN of force. If everything weighs even a ridiculously high 2500 kg with the mass of gas plus the thruster system, the lift from the thrusters is >3x the mass of the vehicle. When you have more lift than mass, you can get off the ground, when it’s that much more over time you have spare air to flip and manuever - simple math really….

Tesla's Q3 earnings deck: "Optimus production began in Fremont on our high-volume lines. The bots coming off this line are the first production-quality bots that we have manufactured. While we will continue to iterate on the hardware and manufacturing process, we have proven that the current manufacturing process for Optimus is suitable for a commercial product."

Most people just saw a humanoid robot throwing kicks and flipping off cars. But the H2-X has some technical details worth looking at. Here’s what the footage and the available specifications tell us about Unitree’s Kung Fu Legend Guardian. ▪️ Height: approximately 170 cm, according to the H2-X booth panel. ▪️ Weight: approximately 65 kg, according to the same panel. ▪️ Degrees of freedom: 31, covering the robot’s articulated body. ▪️ Joint layout: 6 DOF per leg, 7 per arm, 3 at the waist and 2 at the head. This breakdown is reported for the H2 and has not been independently confirmed for the H2-X. ▪️ Battery endurance: Unitree advertises approximately 8 hours. The operating conditions and runtime during intense movement have not been published. ▪️ Battery design: two battery modules are visible on the robot’s back. Their capacity has not been disclosed. ▪️ Joint torque: figures of up to 360 N·m for the legs and 120 N·m for the arms are associated with the H2. They are not confirmed H2-X specifications. ▪️ Computing power: up to 2,070 TOPS is reported for a high-end H2 configuration. Its use in the H2-X remains unconfirmed. ▪️ Sensors: binocular wide-angle vision, an inertial measurement unit and joint sensors are listed for the H2. The H2-X sensor package has not been fully documented. ▪️ Speed and payload: figures below 2 m/s, 7 kg continuous payload and 21 kg short-duration payload are reported for the H2, not confirmed H2-X ratings. ▪️ Recovery: the demonstration shows the robot getting back up after a vehicle runs over it. No standardized recovery test results have been published. ▪️ Autonomy: no detailed explanation of teleoperation, preprogrammed sequences or onboard decision-making has been released. One detail deserves attention: the booth lists the H2-X at around 170 cm and 65 kg, while many reports quote 182 cm and 70 kg from the older H2. The distinction matters when comparing the hardware. Unitree has yet to publish a complete H2-X datasheet covering torque, computing hardware, payload, battery capacity and autonomous operation. Which specification would you want Unitree to publish first? #Unitree #HumanoidRobots #Robotics #PhysicalAI #AI

Why is nobody talking about this? ToddlerBot is a 0.56 m, 3.4 kg humanoid robot built at Stanford by Haochen Shi, Weizhuo Wang, Shuran Song and C. Karen Liu. Its goal is to make robot learning cheap and easy to reproduce. The full hardware costs under 6,000 USD. Hardware - 30 active degrees of freedom: 7 per arm, 6 per leg, 2 in the neck, 2 in the waist - Dynamixel servos (XC330, XC430, 2XC430, 2XL430 and XM430 in the current open-source configuration) - 3D-printed structure: a full repair takes about 21 minutes of printing and 14 minutes of assembly - Jetson Orin NX 16 GB, two fisheye cameras, an IMU, two microphones and a speaker Sim-to-real - Zero-point calibration and motor system identification produce a digital twin of the robot - Walking: an RL policy trained in MuJoCo MJX and transferred to the real robot with no extra training, up to 0.25 m/s in any direction and 1 rad/s in place - Push-ups and pull-ups: keyframed motions, also transferred with no extra training Manipulation - RGB diffusion policies trained on 60 teleoperated demonstrations per task - Real-time onboard depth estimation at 10 Hz from the stereo fisheye cameras - A policy trained on one robot runs on another, and two robots tidied a room together Measured limits - Payload: 1,484 g, about 40% of its body weight - Endurance: 19 minutes of walking before motor heating affects stability - Durability: up to 7 falls before a part breaks - The cartwheel works, but its success rate is still low

This might be one of the most minimalist snake robots ever built. • One motor. • No dozens of actuated joints. • Crawls, climbs between walls and swims. It's called SAW (Single Actuator Wave-like robot), developed by Dr. David Zarrouk's Bio-Inspired and Medical Robotics Lab at Ben-Gurion University of the Negev. How it works • A single motor rotates a helix inside the robot's body. • 3D-printed links follow the helix to create a continuous wave. • The links push against the surrounding surface to propel the robot. • Front wheels handle steering. The numbers • 57 cm/s on flat ground • Around 5× faster than similar wave robots • Around 8 cm/s in a tunnel, with contact against both walls • Around 6 cm/s while swimming in its waterproof version One gait, multiple environments • Gravel, sand and grass • Obstacles and narrow passages • Pipes, tunnels and water The same wave-like motion works across these environments without changing the gait. Why use one motor? Traditional snake robots often rely on 10, 20 or 30 motors. SAW uses a single actuator, reducing the number of moving parts and making the design lighter, simpler to manufacture and suitable for 3D printing. Possible applications • Search and rescue in collapsed buildings • Inspection of pipes and tunnels • Medical imaging and biopsies inside the digestive tract, with a miniaturized version around 1 cm wide or less The research was published in Bioinspiration & Biomimetics under the title Single actuator wavelike robot (SAW): design modeling, and experiments. What caught my attention is how the helical mechanism generates the motion with just one motor. The mechanical design does most of the work

What does it take to build a humanoid robot that can walk without falling? Dropbear recorded 128 walking tests with zero falls. But every test ran in simulation, with external pushes turned off. Developed by Hyperspawn and Point Blank, Dropbear is an open-source humanoid robot. The entire robot is available on GitHub, giving developers access to the project and a chance to study how it works. The next step is testing how well it handles disturbances and unexpected changes in the real world

Unitree Introducing: H2-X Kung Fu Legend Guardian🤖 Powerful performance to serve and guard you, with a maximum endurance of approximately 8 hours. Please use the robot safely and in a friendly manner. https://t.co/5RU5kDzSUK

This might be how machines learn to walk blind. No cameras. No LiDAR. The Noetix N2 (1.2 m, 33 kg) climbs 20 cm curbs using feedback from its joints and an IMU that acts like an inner ear. Every 20 ms, a recurrent neural network reads how the body moves and adjusts its next step based on what it feels. The policy was trained in simulation in just 5 hours on a single GPU. On the real machine it handles: • 15 cm steps • 20° ramps • 20 cm curbs • Uneven stone paths The system relies on proprioception, sensing the body's own movement instead of looking at the terrain. What stands out to me is how much information the joints provide when vision is unavailable.

Spaculation: Might Digit V5 use LFP batteries? V5 has an unusually high 10:1 run-to-charge ratio: 90 minutes of runtime and only 9 minutes to recharge. Few battery chemistries are built for that charge rate. LFP is one of them. LFP cells are heavier than most chemistries, but they tolerate 0–100% state of charge, support fast charging, and are less prone to thermal runaway and fire, making them a safer fit for a safety-conscious company. Assuming a 48 V architecture (higher voltages raise safety concerns, and Agility emphasizes safe operation) and a nominal 1 kW operating load. 90 minutes of runtime then implies a 1.5 kWh pack. Recharging that pack in 9 minutes would take about 208 A. Current at that level typically needs liquid cooling, something the robot may already have for high-end compute and sensors. Combined with LFP cells, that would also help explain the large backpack and the extra weight (285 lb). thanks to Jordan @LimitingThe for suggesting LFP as a possibility in our exchange early this morning

Your phone probably costs more than this humanoid robot. Zeroth 01 is an open source humanoid 40 cm tall with a parts list starting at $350. Most of the frame is 3D-printed Feetech servos drive the joints and you control it in Python. One developer already built one with 16 servos a Raspberry Pi 4 and a battery. It can kneel wave and do push ups. Walking is another story. Five walking controllers have been trained in MuJoCo and none of them runs on that physical robot yet.

‣ 20 Predictions That Will Change How You See Robots Forever ➤ 1. More robots than cars ➝ Billions of humanoids in use, often several per household. ➤ 2. Elder care becomes the biggest market ➝ Aging populations make assistance, mobility help and companionship the main use. ➤ 3. Human-level dexterity ➝ Ultra-sensitive hands that can handle an egg or a surgical tool. ➤ 4. Robots as a subscription ➝ People rent "robot labor" by the hour or month instead of buying the machine. ➤ 5. Days of battery life ➝ New batteries and wireless charging floors. ➤ 6. Robot law ➝ Rules on liability, insurance and privacy, plus debate over legal status. ➤ 7. Nearly human robots ➝ Realistic skin and faces in hospitality and care, mandatory mechanical look in industry. ➤ 8. Work and taxes reinvented ➝ Robot taxes, shorter workweeks or universal basic income. ➤ 9. Robots building robots ➝ Factories, mines and construction sites run almost entirely by humanoids. ➤ 10. A space workforce ➝ The first builders on the Moon and Mars. ‣ Part 2: 10 New Predictions ➤ 11. Mind control ➝ Brain-computer interfaces let people pilot a robot remotely, like a second body. ➤ 12. Artificial muscles ➝ Motors give way to soft fibers for silent, fluid movement. ➤ 13. A tutor for every student ➝ Teaching robots adapt to each child's pace and support human teachers. ➤ 14. Robot rescuers and firefighters ➝ First on the scene in fires, earthquakes and hazardous zones. ➤ 15. Precision farming ➝ Fruit-by-fruit harvesting, plant care and fewer pesticides. ➤ 16. Robot sports and entertainment ➝ Sports leagues, stunt doubles and robot actors become common. ➤ 17. Custom personalities ➝ Voice, character and appearance customizable like an avatar. ➤ 18. Cybersecurity becomes critical ➝ Hacking a robot becomes a serious crime, with mandatory security certification. ➤ 19. Repairable, recyclable robots ➝ Modular design: you swap a part, not the whole robot. ➤ 20. Universal translators ➝ Every robot speaks every language, serving as guide and interpreter anywhere. ✓ Key Takeaways • These are educated guesses, not certainties. • In 75 years, unforeseen breakthroughs (or setbacks) could change everything. What do you think of these predictions?

How to build a humanoid robot at home

You can build a dexterous robotic hand for just $314. Prefer it ready to use? The fully assembled version costs about $720. Chestnut Robotics' Aero Hand Open weighs 389 g and has 7 active degrees of freedom across 16 joints, 3 of them in the thumb. Here's the trick cables run through the fingers like tendons so one motor bends several joints at once. Everything is open source CAD STEP files, 3D-printable parts, firmware and a MuJoCo simulation model. It's still a research prototype and its long term reliability hasn't been proven yet.

Alef Aeronautics, the Silicon Valley startup behind the world's first street-legal flying car, has threw down a cheeky challenge to @Tesla 🔥 The company dropped a video showing its all-electric Model A lifting straight off the pavement and gliding smoothly over a lineup that included a Cybertruck and a Tesla Roadster. The timing could not have been sharper, landing the exact day Tesla showcased its next-gen Roadster. More than just a viral stunt, Alef confirmed the liftoff marked a massive milestone: its 1,000th recorded flight. CEO Jim Dukhovny greeted the newcomer with a memorable one-liner: "Welcome, Tesla, to the flying car race; we've been here alone for eleven years." He added that heavyweight competition is always a win because it validates consumer demand and sets an industry benchmark. What makes this rivalry fascinating is how differently each team plans to conquer gravity. Alef relies on distributed electric propulsion (DEP), using ambient air and internal propellers to hover and transition into forward flight. Tesla, meanwhile, plans to blast the Roadster airborne using cold-gas thrusters powered by compressed air. Yet beneath the surface, both contenders share surprising common ground. Each vehicle runs on an all-electric, all-wheel-drive powertrain, and both secured pivotal early backing from venture capitalist Tim Draper. Flying cars were supposed to stay trapped in retro sci-fi movies. Now, an agile startup and the titan of EVs are taking to the skies with completely different playbooks, and the race to own the air just shifted into high gear!

I've spent years in China building robots. I can tell you this playbook won't work: Pre-seed $: Invent the ultimate actuator. Seed $$: Figure out mass production. The hard part isn't inventing a beautiful topology. It's process engineering. The real delta between the US and China is the ability to execute the enormous amount of physical x mental work required to iterate thousands of times on product designs and manufacturing processes. Partly because much of the tacit knowledge / management systems has left the US, but also because of the sheer human labor required. This isn't something you can solve by buying traditional automation lines either. It requires: 1. Ideation driven by human intuition. Walking up and down the line, talking to assembly workers, and developing hypotheses about what and where improvements can be made. Raw human hours and attention spent understanding a specific manufacturing process. 2. Large batches of experiments across multiple variables. Coordinating experiments that can easily involve 50+ people, issuing instructions, and tracking the execution and results of each experiment. 3. Purpose-built test design to identify key metrics. For example, a change in gear hobbing profile might require accelerated aging tests on a dyno, while a change in bearing tolerance might require NVH testing. 4. Data + human intuition-driven analysis. Interpreting test results and deciding whether to implement a process change or run the next set of experiments. 5. Strict process control. Reissuing work instructions to the line and ensuring every process is followed to the letter. The best manufacturers in Asia, from Foxconn to TSMC, have built their competitive advantage on systematically turning thousands of physical experiments into better manufacturing processes. Their moat isn't just production scale. It's the organizational machinery that makes manufacturing learn and improve faster than anyone else. I don't think most people in SF clock this. Say you're manufacturing actuator gears in-house in the US. After you've made your first article, will you hire a team of 10 process engineers to coordinate the 50 technicians running concurrent experiments with designs, processes, and materials? To understand wear patterns across every pin in a cycloidal drive after 10,000 hours? To reduce backlash from 15 arcmin to 5, then to 1? Just to keep pace with China's progress over the next few years? Do you have the skill set and management systems in place to do this? That's why our approach isn't just to automate manufacturing. It's to automate the process of learning how to manufacture better. Automate simulation and design optimization. Automate assembly and physical testing. Automate the design and execution of process experiments. Feed real-world experimental data back into simulation to continuously improve sim-to-real accuracy. The goal is to turn what currently takes 50 engineers coordinating thousands of experiments into a closed-loop system that continuously learns and improves. Then the same robotic line mass-manufactures the product. PS: Coming back to actuator design, many actuator designs are already commoditized in China. The project in the video is open source, and CNC machining cost in China is just $27 (watch the video). I hope people understand this reality. I'll link more examples in the replies.

Why does every joint of this $5,000 humanoid hide a cycloidal gearbox? Because 3D-printed plastic can't take concentrated loads. Spreading the force over many contact points is what makes it work. Full breakdown of Berkeley Humanoid Lite https://t.co/iVnL54aVpK https://t.co/F61w2tpGt8

A robotic arm that learns physical tasks by watching you demonstrate them. Meet the SO-101, an open-source robotic arm supported by Hugging Face's LeRobot. ‣ 3D-print the parts and assemble the arm ‣ Demonstrate a task using the leader arm ‣ Record movements and train a neural network with LeRobot ‣ Let the follower arm perform the task on its own What catches my attention is how accessible this makes robot learning. You can experiment with imitation learning on real hardware without building an entire robot from scratch.

@Sentdex @asimovinc add cold gas thrusters

🚀RP1 is officially here. 🌍Putting control of robotics back in developers’ hands. Introducing RP1 (ROBOTO 01) by RoboParty—a high-performance, fully open-source humanoid robotics platform for frontier robotics research and embodied AI model development. 🤖 Open across the full hardware and software stack: • Robot platform: mechanical CAD, bill of materials, and assembly documentation • Actuators: in-house ROMOMO joint designs and related technical resources • Control: electrical systems, firmware, low-level drivers, and motion control implementations • Development: simulation models, SDKs, data processing tools, and training tools • Engineering foundations: technical documentation and research materials Open the black box. Understand the system. Make it your own. A deviation in motion can originate in the model, control policy, actuator response, or mechanical structure. RP1 opens up the design and implementation of these layers so developers can trace problems, modify the system, and validate changes on real hardware. Open-source algorithms for advanced motion control RP1 has demonstrated continuous walking, recovery from disturbances, getting back up after a fall, and human motion tracking. UFO and MimicLite connect motion learning, control policies, and execution on real hardware. The open toolchain connects data collection, simulation training, hardware deployment, and experimental validation—supporting joint iteration across models, control, and hardware. From RPO to RP1, we’re turning one team’s engineering experience into a starting point for many more. Opening the future of embodied AI to everyone ready to explore it. 🔗RP1 on GitHub: https://t.co/kMqFLi3eJ3 #RoboParty #RP1 #OpenSourceRobotics #EmbodiedAI

BREAKING: Marc Andreessen just drew the clearest map he's ever given of where the next trillions actually land. And it isn't the software app layer. His thesis: manufacturing jobs from the last 25 years are permanently gone. Stop chasing them back. The real asymmetric bets are embodied AI hardware (robots, drones, autonomous vehicles, defense) and AI attacking the three sectors where prices have only gone up for decades: housing, healthcare, education. The moat in those sectors isn't technology. It's regulation. Which means whoever breaks through first wins everything. He specifically calls out California's SB 1047 as the template for what nearly happened: a bill that would have imposed sweeping liability on AI developers and chilled the entire buildout. It failed. But the next version is coming. Founders still optimizing for software-layer apps are playing a smaller game than they think. The question isn't whether AI reshapes the physical world. It's whether American regulation hands that opportunity to someone else first. Where to place the bets for the next 18 months: https://t.co/yZxU440dQj

The internet needs to see this robot. Meet Tiao Tiao, a crab-shaped robot with 22 tactile servos. It can walk jump and grab objects. Its compact design makes it a fun project to watch for anyone interested in robotics. https://t.co/bd9g9TP62c https://t.co/mfdwh2Iizm

the cut is doing a lot of work. An unbroken take with a bad first pull tells you more than a clean open on an empty drawer. Recovery is the part that actually generalizes. https://t.co/A9MnFUnJcV

Here’s the full Tesla Roadster unveiling event. 0:15 – Crazy Tesla + SpaceX collab preview 2:25 – Last manually driven Tesla car 2:53 – Taking it way past the limits 3:15 – Countdown & first flight works 5:38 – Flying cars are real 5:41 – Peter Thiel flying cars quote 6:02 – Fan suction system (up to 2G) 6:16 – Downforce & crushing track records 7:08 – Starman appearance 7:52 – Rocket thrusters & full gymnastics 8:52 – Kickflip demo (Tony Hawk approved) 9:32 – Real road car + rocket car future 9:56 – Air tanks, nozzles & specs 10:14 – Rechargeable ultra-high pressure air 10:36 – Car reveal & aerospace design 12:50 – Fan modes 13:37 – Variable pitch rear wing 13:50 – 0-60 goal under 1 second 14:21 – Acceleration demo 16:22 – 0.81s 0-60 achieved 16:26 – Production car will be even quicker 17:05 – World’s quickest production car 17:28 – Top speed & Snake River Canyon teaser 18:26 – Mach 0.5+ goal 18:52 – First best flying car + last best driver’s car 19:04 – Production starts 2028

Roadster event livestream https://t.co/G0jU4mWfid

Let me show you an open source humanoid robot you can actually explore and build on. Meet Primo, developed by Alessio Pagliai. • 30 degrees of freedom • 3D-printable design • Walking control developed with NVIDIA Isaac Lab The project gives robotics enthusiasts a way to explore humanoid design and locomotion without starting from scratch.

Brett Adcock is one of humanoid robotics’ biggest hype men. @Figure_robot is valued at $39B; @Forbes says he's worth $23.4B. Nearly two years ago, he said he saw a path to 100,000 robots across two customers in four years. How far has Figure gotten? Our profile 🤖: https://t.co/zUHJbQfUrL

There's still a hope for robotics-specific models https://t.co/I8vKVPwyTS https://t.co/IKie5FWBYH

What if you could build your own humanoid robot? Beijing Dynamics just released Sprite’s mechanical design on GitHub. 31 joints. 95 cm tall. Open CAD files plus URDF and MuJoCo models. One catch the trained controller checkpoint is missing. Would you build on this design or start from scratch?

China will be making a LOT of RAM in 2028 https://t.co/RiBFTe6JIC

🚀RP1 is officially here. 🌍Putting control of robotics back in developers’ hands. Introducing RP1 (ROBOTO 01) by RoboParty—a high-performance, fully open-source humanoid robotics platform for frontier robotics research and embodied AI model development. 🤖 Open across the full hardware and software stack: • Robot platform: mechanical CAD, bill of materials, and assembly documentation • Actuators: in-house ROMOMO joint designs and related technical resources • Control: electrical systems, firmware, low-level drivers, and motion control implementations • Development: simulation models, SDKs, data processing tools, and training tools • Engineering foundations: technical documentation and research materials Open the black box. Understand the system. Make it your own. A deviation in motion can originate in the model, control policy, actuator response, or mechanical structure. RP1 opens up the design and implementation of these layers so developers can trace problems, modify the system, and validate changes on real hardware. Open-source algorithms for advanced motion control RP1 has demonstrated continuous walking, recovery from disturbances, getting back up after a fall, and human motion tracking. UFO and MimicLite connect motion learning, control policies, and execution on real hardware. The open toolchain connects data collection, simulation training, hardware deployment, and experimental validation—supporting joint iteration across models, control, and hardware. From RPO to RP1, we’re turning one team’s engineering experience into a starting point for many more. Opening the future of embodied AI to everyone ready to explore it. 🔗RP1 on GitHub: https://t.co/kMqFLi3eJ3 #RoboParty #RP1 #OpenSourceRobotics #EmbodiedAI