Big bots *look* impressive, but there's lots of room at the bottom.

I predict the first bots used in close combat will be networked swarms of wasp-bots equipped with human-sniffing chemoreceptors and armed with poison darts. Cheap, expendable, difficult to detect or hit, deadly to personnel, and very effective at holding ground.

I remember the DARPA challenge. It was an off-road race for autonomous vehicles. The speeds were pitifully slow and the rigs on the cars were ridiculously cumbersome. The last grand challenge was held in 2005, then they did an urban version in 2007. Now, ten years later we have companies actually using driverless vehicles and trucks on public roads.

I guess what I'm trying to say is that when it comes to robotics and automation and other forms of high tech implementation, the delay between concept and adoption is getting smaller and smaller. The only impediment seems to be making the technology cost-effective, or at the very least, profitible.

You won't see robots replacing front line grunts any time soon. It's a waste of money for standard infantry slots.

Specialized roles though, that's much more likely. A parkour bot would be great on deep recon and the like. Bomb disposal, SAR, sentry, hunting... Missions that can benefit from not needing to stop to eat or sleep.

Huckleberry wrote:

As long as there continues to be a person in the loop, I think we will be able to keep from crossing that line irreversibly.

Huckleberry wrote:

The possibility is there and it is only waiting for the right human weakness at the right time to take us past that line.

I'd say as soon as we let a machine have full control over deciding whether or not to kill a person that'd be pretty much the final step past the "event horizon" of the Technological Singularity. The amazing thing to realize is that most of us will likely be alive to see that happen. How long we all live [b][i]after[/i][/b] that point is another question entirely.

Riptide wrote:

Lothic wrote:

Of course this naturally means the "actual" current state of robot technology could already be far enough along to have functional Arnold-styled terminators that we simply have not seen yet. ;)

Or maybe we see them all the time and don't know they're robots! ;)

[youtube=600x350]1mR9hdy6Qgw[/youtube]

We might be good, and get better, at mimicking artificial intelligence as code gets longer and more complex, but, IMO, no robot will ever get to the point of asking "Who, or what, am I? What is my purpose?" without that being written into the coding. At least, that's my criteria for [i]artificial intelligence[/i].

desviper wrote:

Lothic wrote:

I have a co-worker who has an interesting pet conspiracy theory that all these "advanced robot" vidoes you see on YouTube are actually part of a organized plot to make us think the technology is NOT as far along as it actually is.

The basic idea is that if you ever see anything like Cyclops' attached video what you're looking at is not the "bleeding edge" of what robots are just now able to do in 2018. Instead anything you see in such a "new" publicly released video is likely stuff robots have ALREADY been able to do for like 10-15 years at least. The "gov'mint" doesn't want you to see the super classified stuff so they only let people release stuff to YouTube that's effectively been "declassified".

I must admit the theory does sound appealing in that advanced research organizations (i.e. DARPA, etc.) are not likely going to want to publicly release videos of super top secret cutting edge stuff on YouTube. Of course this naturally means the "actual" current state of robot technology could already be far enough along to have functional Arnold-styled terminators that we simply have not seen yet. ;)

My dad's like this, esp with sattelites. He thinks they can see your face from space. I tried to explain the physics of it :p

All joking aside I can publicly say that I work for a US government contractor that does radar operations and maintenance (O&M) work for several of our Missile Range Instrumentation Ships. I have worked aboard several of these (e.g. [url=https://en.wikipedia.org/wiki/USNS_Observation_Island_%28T-AGM-23%29]USNS Observation Island[/url], [url=https://en.wikipedia.org/wiki/USNS_Invincible_%28T-AGM-24%29]USNS Invincible[/url] and [url=https://en.wikipedia.org/wiki/USNS_Howard_O._Lorenzen_%28T-AGM-25%29]USNS Howard O. Lorenzen[/url]) over the last 15+ years.

I mentioned my work for the following reason: Even if relatively tiny satellites can look down and see things like your dad says (which I am neither confirming nor denying) then imagine what some of the most powerful ship-based phased-array radars on the planet can do when they look [b][i]UP[/i][/b]. ;)

desviper wrote:

I mean radar is a lot easier than telescopy :p

I'm not sure you actually understood what I meant... We don't just "track" things with radar - we "image" things with radar whether they are boosting, travelling at orbital speeds/trajectories or falling ballistically back to Earth. To be more precise mere "telescopy" is trivial compared to what we do and, not for nothing, your dad is not as crazy as you think. ;)

Planet10 wrote:

desviper wrote:

My dad's like this, esp with sattelites. He thinks they can see your face from space. I tried to explain the physics of it :p

Did he happen to watch Enemy of the State by any chance?

Yeah being able to "read wristwatches from space" is still likely something that only happens in Hollywood movies. Being able to visually watch and cleanly identify (via radar imagery) various missile-based payload objects that are boosted hundreds of kilometers into space precisely separate and maneuver 3-dimensionally around each other as they're flying thousands of kilometers per hour thousands of kilometers away from you is a different story. Remember that for "mutual treaty verification" purposes we're mostly doing these things against missiles that are launched by people who are "not our friends".

Just pointing out that some things that seem "plausible" are still Hollywood BS while other things that seem relatively unlikely are actually quite real enough.

warlocc wrote:

That's why I wear wide brimmed hats.

At least they're not aluminum foil :)

Lothic wrote:

We're not up to the standards of Star Trek type sensors yet but we are pushing things as far as most anyone can in 2018. ;)

The line between "contemporary techno-thriller" and "near-future" marches ever forward. And is none too sharp to begin with. ^_^

Ardrea wrote:

Form follows function. https://fas.org/irp/imint/niirs_c/app3.htm

That was a fun stroll down memory lane. ;)

We still use standardized quality scales/metrics to define the quality of the data we collect. But a lot of it is now post-processed through various sophisticated image analysis tools and officially interpreted by scary-smart people with PhDs who work at the various "[url=https://www.urbandictionary.com/define.php?term=Three-letter%20Agencies]three-letter agencies[/url]". Even stuff that starts off looking like proverbial "garbage" data-wise can often be salvaged to provide incredible insights. ;)

Ardrea wrote:

I have been a contractor consultant for several three-letter agencies, so I don't want to say much here.

Those curious can find a peek into that world, both open and formerly-black, from this link: https://en.wikipedia.org/wiki/Reconnaissance_satellite

Yeah as much as I generally leave out "specifics" from my public posts about my job (numerical system specs, taskings, locations, times, etc.) it usually astounds me how much of that you can easily find on public wiki pages. I've worked with Mission Commanders who've come from "intel" backgrounds and they usually have minor strokes whenever we show them how much of our specific "business" exists on the open Interwebs. ;)