If you ever felt disappointed walking into a humanoid robotics event seeing robots that move like awkward puppets, or by watching demo videos full of staged actions that don’t reflect real-world capability…
Then take a look at the Save The Bots Manifesto. It’s a call to build robots that move with physics, not against it.
"The future of robotics will not just be coded, it will be designed"
DM to continue the conversation.
#Robotics #humanoidrobot #humanoidrobotics
@GoingBallistic5@tom_jiahao
1/ General-purpose robotics is the rare technological frontier where the US / China started at roughly the same time and there's no clear winner yet.
To better understand the landscape, @zoeytang_1007, @intelchentwo, @vishnuman0 and I spent the last ~8 weeks creating a deep dive on humanoid robotics hardware and flew to China to see the supply chain firsthand.
Here's everything we've created + our takeaways about the components, humanoid comparisons, supply chains, and geopolitics👇
Meet Sprout.
Today, we’re releasing a new kind of robotics platform. One designed to move out of the lab and into the real world, closer to the people who will shape what robots become next. [1/4]
@GoingBallistic5 Impressive recovery indeed! @GoingBallistic5 would you be more excited in seeing a partial recovery: contact with the floor (one knee or hands), stabilization, then standing up again?
I think that kind of controlled motion would tell a lot about safety margins in real-world.
NEWS: Boston Dynamics has just released a new video of its upgraded next-generation humanoid robot called Atlas.
• 4 hour battery. Self-swappable for continuous operation
• 6 feet 2 inches tall
• Weight: 198 lbs
• 56 total degrees of freedom
• Now fully electric, ditching older hydraulic systems
• New lightweight mix of aluminum and titanium components
• 110 lbs weight capacity (66 lbs sustained)
• Can reach up to 7.5 ft
• Constantly evaluates its surroundings and adjusts its posture, balance, and grip in real time
• Hands that can reconfigure as needed. Tactile sensors feed data back into the system, helping apply the right amount of force
• Brain is powered by Nvidia chips
@GoingBallistic5@chris_j_paxton Agree: performance, reliability and safety overtime have not been showcased. Not even by one company. We are still at the prototyping phase, trying to define the right architecture(s). That's OK, but we must move from the "hardware is good enough" phase.
@MarwaEldiwiny Deeply agree. "good enough" hardware is OK for demo, but not for real life applications. And we need real life applications to improve physical AI. Dumb robots working 20h per day on easy tasks would be very useful for next gen physical AI.
@tom_jiahao I like that indeed. And what you wrote also about biomimicry is important: being inspired by nature is great, but only if the engineering process is anchor in a first principle approach.
Looking forward to your next demo!
Here is the fundamental switch pushed by @tom_jiahao (quote from his article): "Instead of minimizing degrees of freedom or simplifying hardware to make control easier, we take the opposite approach. We give the system as much actuation, redundancy, and sensing bandwidth as practical, then let artificial intelligence exploit that expressiveness. A high dimensional hand has more ways to achieve a goal, more ways to correct errors, and more ways to adapt to new tasks without mechanical redesign. In a world where control algorithms are improving quickly, the opportunity cost of under building the hardware is enormous. The real risk is not in making something too expressive, but in making something that artificial intelligence will eventually outgrow. Our view is that the winning platforms will be the ones whose hardware expands the possibility space for intelligence rather than constraining it." 👌
Wrote a post with some hot takes on dexterity
• Tactile sensing is valuable but misunderstood
• Vision compensates for more than people expect
• Expressive hardware gives AI more room to work
• Biomimicry rarely survives physics
https://t.co/cn5lPwRgWq
Wrote a post with some hot takes on dexterity
• Tactile sensing is valuable but misunderstood
• Vision compensates for more than people expect
• Expressive hardware gives AI more room to work
• Biomimicry rarely survives physics
https://t.co/cn5lPwRgWq
@MarwaEldiwiny@GoingBallistic5 Great video! There is always valuable insight to take away from your content, and this interactive format adds even more. Thank you for sharing the recording.