正在加载视频...

视频加载失败

Sanctuary AI announces its latest generation hydraulic actuation hand, capable of precise manipulation. Hydraulics have significantly more power density than electric actuators in terms of force and velocity. Sanctuary claims to have invented miniaturized valves that are 50x faster and 6x cheaper than off-the-shelf hydraulic valves. The company also...

112,482 次观看 • 1 年前 •via X (Twitter)

10 条评论

The Humanoid Hub 的头像
The Humanoid Hub1 年前

I got to chat with the interim CEO of Sanctuary, James Wells, at the Humanoids Summit yesterday. He was really confident about their ‘hands-first’ development roadmap and their ability to leverage their IP portfolio.

Phil Trubey 的头像
Phil Trubey1 年前

I do like Sanctuary’s focus on hands since that is one of my three key techs needed for humanoids. However their claim that hydraulics are faster than electric “per unit volume” of actuator is a useless factoid. Tesla gets around electric actuator bulk by putting them in the forearm and using “ligaments” to control the fingers. Meaning, there is more to engineering than focus on one metric. I’ll be betting on electric actuators to win.

The Humanoid Hub 的头像
The Humanoid Hub1 年前

They are focusing on heavier-duty high-force manipulation. Packaging and miniaturization need work for sure.

Oli 的头像
Oli1 年前

look really primitive and bulky compared to teslas or 1x design and with less actuators so less control and the massive hydraulic hose isnt helping and the hydraulic power density thing just isnt true anymore even boston dynamics got rid of hydraulics electric is just better now

The Humanoid Hub 的头像
The Humanoid Hub1 年前

Hydraulic actuators definitely generate significantly more force per unit volume, but there are other factors that favor electric actuators.

Brandon Rasman 的头像
Brandon Rasman1 年前

Didnt Boston Dynamics scrap their hydraulic robots for electric motors after years of using hydraulics?

Albert Caro 的头像
Albert Caro1 年前

Im pretty sure that hydraulics are too expensive to make sense when compared to electric actuators. I say this because I imagine the cost of having to assemble and disassemble the components every time it begins leaking. Hydraulic seals don't last very long under perfect conditions. Now imagine the hand working with abrasive dust and all that getting on all those tiny little seals. I must be missing something because it seems too finicky to own after 5 years. Sure you can get much higher quality seals that won't oxidize as quick but you are still looking at massive overhaul costs periodically. Idk that's my shitty opinion on that

Free 的头像
Free1 年前

Hydraulics seem like a backwards step.🤔

Ben Pielstick 的头像
Ben Pielstick1 年前

Maybe @BostonDynamics changed directions too soon.

The Humanoid Hub 的头像
The Humanoid Hub1 年前

@BostonDynamics The rest of the Phoenix body moves through electromagnetic actuation. Only the hand is hydraulic actuated

相关视频

Ron DeSantis says the Congress must cut off all federal funding to Democrat states who have Sanctuary Cities to force them to change policies “I just hope that the Congress, for once, will use the power of the purse and defund sanctuary states and defund sanctuary cities. Money talks, and if you do that, you'll change the behavior” “When Biden was president, he opened the border. States like Florida and Texas, we were actually passing laws to do immigration enforcement since he wasn't. And what did the courts say? "Oh, no, you can't do that. It's federal." Then flip it, Donald Trump closes the border. He's now trying to undo the damage that President Biden did, and these places are sanctuary, and the courts are, instead of saying that they're violating the law by trying to frustrate federal law, they're trying to throw sand in the gears of the Trump administration. Almost anything they do, you'll have some liberal judge that will try to stop it. And so what the left wants is a one-way ratchet. They wanna be able to open the border, import a new electorate, and then even if there's a reaction and you have a new president come in, they don't want the, that president to have the power to undo the damage that was done. And so we can't let that happen.” Our money is being thrown away by giving it to sanctuary states and this is a bigger issue than you realize $800B–$1T+ is given to sanctuary states every single year. If you were to cut this off, every single one of them would instantly change their policies Like Ron DeSantis says “Money talks”

Wall Street Apes

41,866 次观看 • 8 天前

The Chinese are advancing on 6th-generation fighters, but I have serious doubts about whether this is truly the right path. Recently, the Turkish company Baykar flew two combat drones that operate in perfect synchronization, guided from a ground base. They executed coordinated maneuvers autonomously and are already capable of conducting BVR (Beyond Visual Range) engagements. The Kızılelma became the first drone to achieve an autonomous kill with an air-to-air missile guided by its own AESA radar. This level of Turkish progress, combined with the kill web integration that Chinese drones are demonstrating, isn’t it a strong indication that manned 6th-generation fighters might be an extraordinarily expensive fantasy? I’ll go further: some drones are already reaching speeds above Mach 3.5 - more than double those of the leading current manned fighters. I’m talking about drones capable of high-speed coordinated operations, fully autonomous, with integrated and shared weapon systems, flying at altitudes far higher than conventional fighters, and employing systems similar to Aegis. How much longer until these drones take off in large swarms, equipped with ultra-modern sensors controlled by AI and armed with long-range air-to-air missiles? How could a human pilot, even at the height of adrenaline, make faster and more accurate decisions, involving multiple scenarios and factors, than an AI commanding a pack of smaller, more agile drones operating at different altitudes? We are entering the real competition between man and machine. Even if a pilot in a 6th-generation fighter has control over slave drones and an entire Aegis-like kill web, what is the point of having him there if everything can be done autonomously, more safely, more cheaply, and in far greater numbers? Some argue that the human pilot offers advantages in unpredictable scenarios where today’s AI still fails, and I agree. But I’m not talking about current AI: I’m talking about the AI of 2030–2040, far more advanced than anything we know today. Geran drones equipped with air-to-air missiles prove that air combat drones can be extremely low-cost, while the Turkish tests show, in a far more sophisticated way, that drones designed for air-to-air missions are evolving much faster than 6th-generation aircraft, whose development and production costs are exorbitant. When we put together the Geran, the Kızılelma, and the drone kill web integration the Chinese are achieving, the question arises: is the future really manned 6th-generation fighters, or much cheaper drones produced in greater numbers, better protected, and more capable? Wouldn’t it be smarter to invest in well-equipped autonomous 6th-generation drones rather than treating them merely as wingmen? It seems that keeping a human pilot on board has more to do with ethical principles and preserving our role than with genuine operational necessity. And how would humanity react to the constant killing of human beings by AI-guided drones? Drones will be cheaper to develop, easier to produce, and possibly more agile than conventional fighters, even 6th-generation ones. In cost-benefit terms, the price of a single 6th-generation fighter could buy dozens of equivalent or even superior drones. This would save hundreds of billions of dollars and could put whoever adopts the strategy a step ahead. The Chinese, by the way, are very well positioned in this race, with HALE drones already armed with air-to-air missiles and integrated into an extensive kill web, which I will describe in detail soon.

Patricia Marins

60,728 次观看 • 6 个月前

Everything Elon said about Optimus at the All-In Summit today: • We’re finalizing the design of Optimus v3. That release is going to be a very remarkable robot. It will have manual dexterity comparable to a human, meaning a very complex hand, an AI mind that can navigate and comprehend reality, and will be made in very high volume. • Other robotics companies are missing those three very hard things. • I spend more mental cycles on Optimus than any other single thing. Solving real-world AI, all of the electrical-mechanical issues, the supply chain, and production challenges. • There is no supply chain for humanoid robots, so it has to be created from scratch, which requires a lot of vertical integration. None of the actuators in Optimus are available from an existing supply chain. • I think if successful, Optimus would be the biggest product ever. • The marginal cost of production, once we hit a million units per year, will probably be around $20,000. It depends on how much we spend on the AI chip in the robot, and we’ll need to achieve a lot of efficiencies in the actuators—26 actuators per arm (26 motors, gearboxes, and power electronics). The AI chip might cost $5,000 or $6,000, maybe more. At 1 million units a year, production cost will be $20,000, maybe $25,000. Price will be a function of demand. • Human hands have evolved to be incredibly sophisticated machines. Hands are a very first instrument. You can swing a baseball bat, thread a needle, play a piano or violin, and assemble a car. Hands are incredibly versatile instruments. Most of the muscles of the hands are actually in the forearm, and the hand is almost like a puppet. Human tendon evolution is incredibly good. The human hand has 27 or 28 degrees of freedom, depending on how you count it; it’s amazing. • In order to create a robot that can be a generalized humanoid, you must solve the “hands problem.” • Even though there are 10,000 to 20,000 electric motors out there, we couldn’t buy the actuators for any amount of money. We had to design every electric motor, gearbox, and controlling electronics from scratch, from first principles of physics. • Optimus is harder than developing any previous Tesla product, but not harder than Starship. • Right now, we’re struggling with the final design of the hardware, primarily the hand. The hands and forearm are the majority of the engineering difficulty of the entire robot. • If you want to do all the things that a human can do, it turns out you need a humanoid robot. If you want to do a subset, that’s much easier. Humans evolved to the shape and capability that we have for a good reason. There is value to having four fingers and a thumb; even the pinky is quite useful. Toes are much more of a question mark. • The AI5 inference chip will be 40 times better than AI4 by some measures. We know the limiting factors of the chip because the AI software and hardware teams work so closely. Effectively, the Tesla AI hardware and software teams are co-designing the chip. • The Softmax function on AI4 takes 40 steps in emulation mode, which will take only a few steps in AI5 natively. AI5 will easily handle mixed precision. • In terms of nominal raw compute, the AI5 inference chip has 8 times more compute, 9 times more memory, and 5 times more memory bandwidth compared to AI4. Because we’re addressing some core limitations and optimizations at the silicon level, we’re able to realize 40x improvements.

The Humanoid Hub

238,877 次观看 • 11 个月前