I would like to see the last 20 seconds of this video with a Dinosaur Jr soundtrack.
Also I don’t get why you are all convinced this is a warbot. Clearly the benevolent and altruistic Boston Dynamics (you can tell by their name that they are the good guys) are designing this as a mechanical suit to be worn as a replacement for the wheelchair!
Hmm. Has anyone here actually studied robotics or coded robots? Boston Dynamic’s cool robots are Hollywood nonsense. When you build a cost-effective and worthwhile robot you build it not generally but specifically. I can only assume most people’s knowledge of how robots are supposed to work are colored by sci fi fiction like I, Robot and Westworld. Humanoid robots are just terrible at doing specific jobs, it is much, much better to design the robot around the job than to build a general humanoid robot that can do everything. Take for example fast food service robots. Do you think they will look like people with little hats on? It will look like a bunch of big hoppers full of food, chutes, maybe some pivot points and arms. No where will there be anything that looks like a BD robot. You will never see them stacking boxes in warehouses either. Do a GIS of the Amazon warehouse robots if you want to see the future of robotics. The prima facia evidence they are on the wrong track is that statement from Alphabet that mirrors Job in Arrested: “I’ve made a huge mistake”.
These robots are parlor tricks and their code is worthless, IMO. If you want to see the future of robotics watch How It’s Made sometime and pay attention to how they make industrial machines.
That’s quite a leap there.
Ok, maybe not completely worthless. But would you buy it? I wouldn’t.
Buy what, the robot? No, I wouldn’t buy it, but my understanding is that their work is about R&D in various areas of robotics, not trying to create sellable robots. So the fact that I wouldn’t buy it is kind of moot, in my opinion.
If we only ever did work that resulted in “marketable products within a few years” this country would be in deep doodoo. This is the same kind of attitude that drove GM to bankruptcy while a Toyota thrived. The work Boston Dynamics is doing could be amazing/critical for extra terrestial (moon), volcanic, and highly adverse conditions such as Fukushima - once we have better energy density to power their movement. To trash their research because it won’t make a buck in a couple years… I can’t even imagine how shortsighted this is. Legged, not wheeled rovers are the ultimate future of space exploration. Look at the surface of Europa.
Also, the coming sexbot market is going to demand this kind of agility.
Perhaps you’re not entirely up to date, Guap?
There’s now a growing sector of end-user-programmable general purpose industrial robots for assembly tasks in light industry. It’s no longer the case that robots have to be specially designed and programmed for particular roles in particular assembly lines.
That said, of course mobile robots like those created by Boston Dynamics are worthless for industrial applications, and due to their power limitations have few if any practical uses today. But when mobile robot applications do arise, whether for delivery and freight or surveillance, or whatever, someone will be the market leader and someone will have all the relevant patents. And it won’t be all that long before the demand for the robots is there, and before the robots have power sufficient for their needs.
And if their power isn’t sufficient, they can eat our medicine for fuel.
Robots are strong.
I admit a bit of armchair quarterbacking but I’ve never heard this. When studying the auto industry it’s much more a narrative of the structure of their part suppliers, quality control, and coordination. There is also the idea of large conglomerates in Japan (this was in the 80s and 90s, we’ve caught up to that idea now). I’ve never heard it as a pure research advantage. Do you have any sources or is this speculative on your part?
I agree the world needs pure research but BD is far from that, in my opinion. They are very much applied research striving for a goal. My argument would be they are wasting time going down the wrong path.
An excellent reference.
What’s new, Atlas? (spoiler: backflips!)
The new SpotMini - this is starting to get creepy. I got a distinct Half Life 3 vibe from this guy.
Watching the Atlas video, they are making amazing strides with the balance and agility of these things. In the slow-mo of the backflip you can see all the micromovements post jump as the robot stabilizes itself.
And I think that’s part of what they appear to be doing there in Boston, progressing movement and balance of different robot types. Reddit mentioned they were bought off Google, so new ownership again?
Ugh, just what we needed, robots pressuring us to join CrossFit.
Yeah, they’re owned by Softbank now. IMO the SpotMini received the attention of a consumer products designer (either that or a mecha-loving nerd engineer, anyway), as its case is much shinier and sleeker than other models they’ve shown. Two-tone plastic cladding and all that. It looks more like a commercial product than their other prototypes. But possibly it was just done to please their Japanese owners.
They need to sample that actuator sound into the next movie about robots gone bad and taking out humans. I can imagine hearing a few of those walking down a hallway toward your hiding spot would be terrifying.
Also, this older video is pretty crazy:
If you look at the screen you see a readout of the lidar sensor in the unit. I can see why DARPA was putting money into Boston Dynamics for a while.
All i have to say about this nonsense is…
Suffer not the AI to live.
I wasn’t worried when a bearded bully easily pushed a ten pound box out of the bot’s hands.
But if they are doing standing backflips, it’s time for me to open a policy with Old Glory Robot Insurance. Because the metal ones are coming for us.
Boston dynamics really seems to have the brainstem down. I don’t think anyone else is even close.