Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Atlas robot does parkour

blog.bostondynamics.com

Post reply on HN
Print view View original

121–130 of 242 posts

Re: Atlas robot does parkour

#121

Post by btbuildem »

handrous wrote: post #113
narrator wrote: post #28

Earlier quoted context omitted.

The military will just use aerial drone slaughterbots. The Azerbaijanis did this in the recent Armenia/Azerbaijan war and it was a total blowout against 20th century tech. Here's what I'm thinking the good future looks like: Let's say I have a big back yard and would like to have that be a productive farm. AI bot comes over, surveys the land, does soil tests, orders seeds, fertilizer and garden equipment off Amazon.…

> The military will just use arial drone slaughterbots. The Azerbaijanis did this in the recent Armenia/Azerbaijan war and it was a total blowout against 20th century tech. It hadn't occurred to me that drones let despots wage high-tempo modern wars (at least, partially) without the associated risks (of devolved control) and costs (of very expensive training and equipment for your soldiers)—but they totally do. That'…

As with anything else, we will have countermeasures for drone warfare as well.

Re: Atlas robot does parkour

#122

Post by ansible »

mastax wrote: post #33
scythmic_waves wrote: post #16

Earlier quoted context omitted.

I'm not so sure. 1. [EDIT: IGNORE THIS POINT] They've done this before. [1] 2. To get more specific, look at the robot's legs around 0:09-0:16. They don't seem to be moving normally. I don't have professional video analysis chops, but I'd put some money on this being fake. Happy be to be proven wrong, though. EDIT: It appears I remembered this story incorrectly, and posted a link without reading it. I retract my firs…

That CGI gag was done by Corridor Digital, not Boston Dynamics.

That's a great channel, I really like their "reacts" series that comes out every Saturday. They also had a good one from Sunday about the Pentagon UFO videos.

Re: Atlas robot does parkour

#123

Post by djrogers »

handrous wrote: post #113
narrator wrote: post #28

Earlier quoted context omitted.

The military will just use aerial drone slaughterbots. The Azerbaijanis did this in the recent Armenia/Azerbaijan war and it was a total blowout against 20th century tech. Here's what I'm thinking the good future looks like: Let's say I have a big back yard and would like to have that be a productive farm. AI bot comes over, surveys the land, does soil tests, orders seeds, fertilizer and garden equipment off Amazon.…

> The military will just use arial drone slaughterbots. The Azerbaijanis did this in the recent Armenia/Azerbaijan war and it was a total blowout against 20th century tech. It hadn't occurred to me that drones let despots wage high-tempo modern wars (at least, partially) without the associated risks (of devolved control) and costs (of very expensive training and equipment for your soldiers)—but they totally do. That'…

There’s already a drone/anti-drone arms race going on, with tech like microwaves, jammers, nets, and other more classified means of drone flight denial.

The drones will be effective for a while, then it’ll become impossible to fly them in airspace controlled by a 1st or 2nd tier power, and you’ll be back to needing foot soldiers (if only to take out the drone jammers).

Re: Atlas robot does parkour

#124

Post by BashiBazouk »

jvanderbot wrote: post #35

This is awesome, but until someone solves the real-time sensing and modelling problem, it'll never fly in the real world. The use case is still warehouses. Here's my logic / assumptions: How does the robot "know" (have sufficient model information to predict the response) that the board will hold it when it does a vault? It doesn't -- it is told to vault by placing one limb on top. As we've seen with self-driving car…

Whenever I see the four legged dog robot, I wonder how far we are from the chevaline in Neal Stephenson's The Diamond Age.

Re: Atlas robot does parkour

#125

Post by habitue »

ClumsyPilot wrote: post #56
freediver wrote: post #26

What makes this really interesting is that all this is achieved with very little or no "AI" (in the context of current machine learning approaches). Purely heuristics based approach to something as complex.

There is weired tendency to use AI for well-understood problems that have formulas and solutions.

Right, but is parkour a well-understood problem with formulas and solutions?

I think there's also a tendency to go "You just need linear regression" a lot when actually, well, there is a reason people are excited about new AI techniques: they make things that were very difficult much easier, and they make some things possible that never were before.

Is deep learning / reinforcement learning actually useful here? Maybe. I would be surprised if Boston Dynamics wasn't at least looking into it.

Re: Atlas robot does parkour

#126

Post by djrogers »

lifty wrote: post #97
mrfusion wrote: post #72

Earlier quoted context omitted.

> We need to decentralise governance and power in a massive way Agreed but I’d also like to see human rights protected from the new decentralized governments. Not sure how that could work.

We could mandate that all AI robots should load an abide by a digital constitution that has been cryptographically signed by a majority of the population.

And how do you enforce your mandate? Without the threat of physical violence, any mandate devolves into a suggestion - or worse, handcuffs on the just.

Re: Atlas robot does parkour

#127

Post by yabones »

prepend wrote: post #18

What can I do to defend myself against killer robots? Seriously, is there something simple that I can put in my glovebox of my car to use in case these robots start getting scary? Is there a simple household good that will work well against these, until they start explicitly shielding against stuff?

I'm going to assume that most 'killer robots' will be reasonably heavily armoured, so unless you have very powerful AP rounds the best approach will be to disable rather than destroy. They'll also likely be hardened against 'jammers' or any radio interference.

Presumably most systems will use multiple sensors and vision systems, LiDAR, RADAR, etc. The key would be to disable enough of each to put the machine in 'safe mode'.

Vision systems are the easiest to attack. A cheap paintball gun, spray can, or fire extinguisher would be more than capable of disabling a camera system.

LiDAR would likely also be fairly easy to disable. Most LiDAR systems also depend on visual light, and have a revolving sensor of some sort (though there are some fixed units as well) which have their own weaknesses. A reasonably powerful pellet gun, or most rifles or pistols would be enough to disable a revolving lidar sensor, and a few shots with paint would disable a fixed unit, much like a camera.

It's almost certain the designer of such robots will think of these attacks and build around them. Think windshield wipers or compressed air to clear debris. So, the ideal material would be very thick, very sticky, and completely opaque. It would also need a high pressure container and a nozzle that produces a laminar stream to hit the sensors from at least 10m away.

Re: Atlas robot does parkour

#129

Post by sangnoir »

lifty wrote: post #97
mrfusion wrote: post #72

Earlier quoted context omitted.

> We need to decentralise governance and power in a massive way Agreed but I’d also like to see human rights protected from the new decentralized governments. Not sure how that could work.

We could mandate that all AI robots should load an abide by a digital constitution that has been cryptographically signed by a majority of the population.

Who will enforce this mandate? Or will this be an honor system?

Re: Atlas robot does parkour

#130

Post by btbuildem »

aggie wrote: post #53
delgaudm wrote: post #8

I'm not a particularly pessimistic person in general, but I don't know ...this tech, while cool, feels like the precursor to a worse future in general. Maybe I've seen too much sci-fi, but these robots look like optimizing for future soldiers, law enforcement etc. Given all the other Self-Driving / Facial Recognition / Loss of Privacy / Deepfake / "AI is bad at recognizing $situation"-type posts, and as all these tec…

"AI Robot cops show implicit bias resulting in 3 deaths" is a headline today if you remove the AI robot part. As evidenced by the comments in this thread and any other on these robots, the public is extremely skeptical of them. If they do indeed exhibit these kinds of harms, the public will likely enact severe restrictions on them. I wouldn't be surprised if we end up in a situation where a lot of benefits are left o…

The robots will just become drones (remotely controlled by humans). BD demos are great to show that these will be autonomously ambulatory (you just tell them which direction to move, they manage the low-level concerns of the movement).

We won't trust AI with decision-making, but we will trust the status-quo police forces to dish out violence from the safety of an office.

Post reply on HN