Any one's tried LLM on a bipedal robot? I feel like with Anthropic's Computer Use (LLM trained with visual and controls a mouse), there start to be some overlapping
Boston Dynamics robot Atlas goes hands on [video]
201–210 of 220 posts
Re: Boston Dynamics robot Atlas goes hands on [video]
#202Earlier quoted context omitted.
The entire built environment to date is designed for humans It’s the same reason why companies are throwing money at a self driving car which can coexist with manually driven cars, rather than building roads specifically for autonomy
Why build autonomous roads? Just invest in modern railways at that point.
Because building anything infrastructure related in the US, especially for rail, is a such a Sisyphean task that it might actually be cheaper to just build an autonomous vehicle
Re: Boston Dynamics robot Atlas goes hands on [video]
#203This would cost as much and more as the lifespan costs of a human in Africa. Anyway, the 3-4-second pause is weird.
Re: Boston Dynamics robot Atlas goes hands on [video]
#204Chinese Unitree's demos are better by now imo. https://www.youtube.com/watch?v=8dFTc4W8wm0
Plus at least 10x cheaper!
Re: Boston Dynamics robot Atlas goes hands on [video]
#205Earlier quoted context omitted.
The entire built environment to date is designed for humans It’s the same reason why companies are throwing money at a self driving car which can coexist with manually driven cars, rather than building roads specifically for autonomy
What's that got to do with vision ? They could have given it more cameras to avoid having to have a head to turn. That was OP's point.
They're responding to the person recommending octopus robots, not the vision comment.
Re: Boston Dynamics robot Atlas goes hands on [video]
#206Earlier quoted context omitted.
How do you know the figure 1 demo was fake?
Motion is the telltale sign. It's too smooth in their video. Real physical mechanisms shake a bit, the actuators have a bit of backlash, the whole structure wobbles a bit when a limb is moved due to torque, etc. The micro-movements are missing from the Figure 01 video, because they are hard to simulate convincingly. Basically, their robot moves like Optimis Prime in a Transformers movie. Unnatural, and un-physical. Y…
Either that or it's really impressive engineering.
Re: Boston Dynamics robot Atlas goes hands on [video]
#207It's a nice piece of machinery. Looks like they finally went all-electric and got rid of the hydraulics. The new model is much less bulky. The task it is doing is undemanding. It's just moving things from one set of large slots to another. No need for precision placement, unstructured bin-picking, or object separation. If it could pick up engine covers from the messy pile seen atop one of the racks and slot them into…
On the shift from hydraulic to electric: for years Marc Raibert and others at Boston Dynamics expressed doubts that a non-hydraulic humanoid could be made to perform as well in that package. Because of actuator density- a humanoid has a high number of actuators. Also power density, and they did a lot of work in molding the hydraulic actuators directly into the arms and legs to make them compact. The bulk is all in th…
> On the shift from hydraulic to electric: for years Marc Raibert and others at Boston Dynamics expressed doubts that a non-hydraulic humanoid could be made to perform as well in that package.
Then Schaft won the DARPA Robotic Challenge in 2013 with their all-electric humanoid robot with liquid-cooled motors. They beat out Boston Dynamics' Atlas. Then Google bought Schaft. Google got bored with robotics, couldn't find a buyer for them, and Schaft was shut down, along with several other Google robotics acquisitions.
Re: Boston Dynamics robot Atlas goes hands on [video]
#208skin reaction: useless! unfortunately the world is full of sooooo many much smarter slaves to exploit
But can you buy them, not pay them, feed them for pennies, keep them alive forever, work them 24hrs/365days, and never fear they will organize and revolt?
yes, that is the scary part you are right, no doubt, but I just don't see it coming very soon (fortunately), and I do really hope I am right
Re: Boston Dynamics robot Atlas goes hands on [video]
#209I don't know why people are impressed. Robots are hard and this is a big step forward but it's still useless. This is far from being as functional as automotive manufacturing robots or even vacuum cleaners. This is nothing like the goal we're striving for with these things. There's very a long way to go until this does what we want.
Why not be impressed? The reasons you gave are enough for the majority of people following along. The exciting parts are the steps along the way. Once we have something that is actually usable and integrated into our everyday lives, it will lose its novelty fast.
> Why not be impressed?
Because it's obviously useless and a regression from the kinds of robots we already have.
Re: Boston Dynamics robot Atlas goes hands on [video]
#210It's a nice piece of machinery. Looks like they finally went all-electric and got rid of the hydraulics. The new model is much less bulky. The task it is doing is undemanding. It's just moving things from one set of large slots to another. No need for precision placement, unstructured bin-picking, or object separation. If it could pick up engine covers from the messy pile seen atop one of the racks and slot them into…
> The task it is doing is undemanding. Have you noticed how in second 13 it rotates the lower body clockwise while the upper body rotates counter-clockwise in order to optimize the movements to archive the goal? All (bipedal/quadrupedal) robots I've seen behave either like an animal or a human, but this one is something else. Imagine the inverse kinematics required to perform such a movement, and how precise even the…
> Also the recovery it had when it failed to insert one of the objects.
You mean the unnecessarily violent jerking motion it does for no great reason at 1:22 instead of gently trying again?