How badly do we need robots? Voice assistants are the first step and while they might be useful for some things they are not widely used. For mundane tasks you have roombas etc, but those fulfill very specific tasks. A versatile robot that can do all your household tasks and complicated things like driving your car are a long, long way off, and i would argue are not even that necessary.
What robots are good at is automation in factories, allowing manufacturers to streamline tasks, perform them faster and save labour, which does drive down prices.
Perhaps with ai the technology to make 'intelligent' robots is finally within sight, but the physical technology is still not there yet, and even if it would be i doubt it would become widely adopted.
We mainly need humanoids to replace jobs with repetitive tasks i.e. harvesting crops or any kind of service related jobs and tasks perofrmed in dangerous environments like mines, disaster zones, war zones etc.
Eventually having a humanoid at home doing the dishes and whatever other tasks we find boring is a byproduct of developing a capable robot for the repetitive and dangerous jobs.
Where's this "preprogrammed" meme coming from? Everything in robotics is still "preprogrammed". Later builds of ASIMO did have dynamic balancing and in fact ran with both feet in the air.
It's not like they just hit the precalculated coordinates on the floor with the feet, they used gyros and hand-written algorithms to compute all the forces in real time, just like today those Unitree bots do with GPU trained algorithms. They're fundamentally the same. Very little had changed.
Who's making up that hallucination? This isn't even the second time I've come across those "preprogrammed" BS.
We've certainly come a long way from Asimo to robots like Boston Dynamic's Atlas doing parkour and backflips, but Asimo was SOTA when it came out - back in the day I took a trip from NJ to Philly to see him being demonstrated at the science museum there, doing things like jogging in a circle or climbing up a set of steps.
Now that the motion part of robotics is pretty much solved, we need grippers and sensors - still being worked on - then enough of a brain to be easily taught new skills and be able to perform them reliably in real-world conditions.
I know I'm going to be downvoted for this, but I have to say that Teslabot shuffling along with what appears to be clenched butt cheeks looks more Asimo era in terms of dynamics than SOTA. It's not even clear if it has dynamic balance or depends on keeping center of mass over its feet.
I still do something I call the "Asimo Walk" (walking with slightly bent knees to take the load off the joints and engage the muscles for support), especially when the tendons around my knee act up.
I saw the Asimo demonstration at the Miraikan in Tokyo (it's not still on but the rest of the museum is great if you are nearby) and I still look back fondly on it today. I saw it too late (2017 and 2019) for it to feel like the future but there was something lovely about how it had a personality and a charm that some of the other robots there.
I remember learning about Asimo in grade school. It felt like the future! Robots would be assisting me in daily life when I grew up.
Then like the space shuttle, it just disappeared. I feel like there was Asimo.... and then nothing for decades until now.
Sure I can have a robot do my dishes, but it's still more efficient to just use a dishwasher appliance.
I’d be happy with a robot that packs the dishwasher, even if that is its only skill
So English not required? LLMs beware.
How badly do we need robots? Voice assistants are the first step and while they might be useful for some things they are not widely used. For mundane tasks you have roombas etc, but those fulfill very specific tasks. A versatile robot that can do all your household tasks and complicated things like driving your car are a long, long way off, and i would argue are not even that necessary.
What robots are good at is automation in factories, allowing manufacturers to streamline tasks, perform them faster and save labour, which does drive down prices.
Perhaps with ai the technology to make 'intelligent' robots is finally within sight, but the physical technology is still not there yet, and even if it would be i doubt it would become widely adopted.
SoftBank bought ABB a week or two back, and they're not saying Dual YuMi will be retailing at Walmart everywhere, so I'd say, no, we're not in a need.
SoftBank bought ABB’s robotics business, not ABB itself.
We mainly need humanoids to replace jobs with repetitive tasks i.e. harvesting crops or any kind of service related jobs and tasks perofrmed in dangerous environments like mines, disaster zones, war zones etc.
Eventually having a humanoid at home doing the dishes and whatever other tasks we find boring is a byproduct of developing a capable robot for the repetitive and dangerous jobs.
Thats not fair, there was AWESOM-O in 2004 :)
Asimo was a dead end. Preprogrammed/precalculated static balance, that uncanny 'Im holding a surprise in my diaper' walk.
Where's this "preprogrammed" meme coming from? Everything in robotics is still "preprogrammed". Later builds of ASIMO did have dynamic balancing and in fact ran with both feet in the air.
It's not like they just hit the precalculated coordinates on the floor with the feet, they used gyros and hand-written algorithms to compute all the forces in real time, just like today those Unitree bots do with GPU trained algorithms. They're fundamentally the same. Very little had changed.
Who's making up that hallucination? This isn't even the second time I've come across those "preprogrammed" BS.
"run" at the speed of brisk walk https://www.youtube.com/watch?v=oV3hvqRtTgM more of a fast sneaking. I dont think I saw real running non tethered robot before Boston Dynamics, and today running is indeed in reach of $20K toys https://www.youtube.com/watch?v=CIkdq7Zf4Zw
We've certainly come a long way from Asimo to robots like Boston Dynamic's Atlas doing parkour and backflips, but Asimo was SOTA when it came out - back in the day I took a trip from NJ to Philly to see him being demonstrated at the science museum there, doing things like jogging in a circle or climbing up a set of steps.
Now that the motion part of robotics is pretty much solved, we need grippers and sensors - still being worked on - then enough of a brain to be easily taught new skills and be able to perform them reliably in real-world conditions.
I know I'm going to be downvoted for this, but I have to say that Teslabot shuffling along with what appears to be clenched butt cheeks looks more Asimo era in terms of dynamics than SOTA. It's not even clear if it has dynamic balance or depends on keeping center of mass over its feet.
Boston Dynamics is now focusing on dexterity stuff for the current Atlas generation's grippy hands. https://youtube.com/watch?v=gS4rOqNDTBk
Interestingly, Boston Dynamics is presently owned by a car company. It's Hyundai rather than Honda.
I still do something I call the "Asimo Walk" (walking with slightly bent knees to take the load off the joints and engage the muscles for support), especially when the tendons around my knee act up.
I saw the Asimo demonstration at the Miraikan in Tokyo (it's not still on but the rest of the museum is great if you are nearby) and I still look back fondly on it today. I saw it too late (2017 and 2019) for it to feel like the future but there was something lovely about how it had a personality and a charm that some of the other robots there.