Wheeled & Non-Wheeled Legged Robots and Ground Vehicles

[sanman]

Here is yet another new variant on the legs-with-wheels approach, but this time it's modular:






So other than doing something because you can, is there some practical reason for modular robots like this?

Using Mars & Moon missions as examples, we can have a big rover acting as a mothership/hub, with smaller machines foraying out from it and returning back to it, perhaps with samples.

But what's the use case for a modular machine like the one in the video above?
If you have a 2-wheeled machine docking with another one, to make a 2+2 / 4-wheeled configuration, then what  advantages are there in being able to do this?

[Paul451]

So other than doing something because you can, is there some practical reason for modular robots like this?

As designed, no. In theory, as end-caps around modular instrument packages / tool systems / containers? Maybe.

Although even then it might be more useful to have a single (solid) quad-leg/wheel base with the click adaptors on top, letting you add an arm(s) unit, and/or sensors/instruments, and/or extra batteries/panels, etc, to adapt to different jobs.

[catdlr]

Here is yet another new variant on the legs-with-wheels approach, but this time it's modular:

https://youtube.com/watch?v=GbL3mjoZtOA

https://youtube.com/watch?v=HLwkS6SGWsY


So other than doing something because you can, is there some practical reason for modular robots like this?

Using Mars & Moon missions as examples, we can have a big rover acting as a mothership/hub, with smaller machines foraying out from it and returning back to it, perhaps with samples.

But what's the use case for a modular machine like the one in the video above?
If you have a 2-wheeled machine docking with another one, to make a 2+2 / 4-wheeled configuration, then what  advantages are there in being able to do this?

There is redundancy. If you go exploring and one unit fails, the other can detach and return to home base. Faults that could occur during exploration include battery drain, getting stuck between boulders, etc. Having a threesome go out on a mission allows two units to reconnect and return to base. Now, if they can get an arm that could lift the dead unit and place it on top, then the remaining unit(s)  can carry it back.

[edzieba]

So other than doing something because you can, is there some practical reason for modular robots like this?

Take one of these modular wheg robots, and reduce the range of motion of the legs from 'lots' (currently) to 'very little other than vertical'. You now have an SPMT. Keep the long legs? You now have a rough-terrain SPMT.

[catdlr]

Here is yet another new variant on the legs-with-wheels approach, but this time it's modular:

https://youtube.com/watch?v=GbL3mjoZtOA

https://youtube.com/watch?v=HLwkS6SGWsY

Here is the American version of that:

https://youtube.com/clip/UgkxmbfFNDt4EPbZQqheTw8rotZCK5lKrQUE?si=gR4P0w8uOcSY1Zdg

[Twark_Main]

Here is yet another new variant on the legs-with-wheels approach, but this time it's modular:

https://youtube.com/watch?v=GbL3mjoZtOA

https://youtube.com/watch?v=HLwkS6SGWsY

Here is the American version of that:

https://youtube.com/clip/UgkxmbfFNDt4EPbZQqheTw8rotZCK5lKrQUE

Original video:

[sanman]

More tricks

[Twark_Main]

Cross-posting from the humanoid thread because it's relevant.

One of the comments that got my attention was the difficulty of unloading the vehicles and setting up various devices from there. Conclusion was that Optimus robots would be required. My reaction was to wonder if that person was aware of current forklift capabilities including some autonomous capabilities.

This applies to pretty much all "use a robot to operate machinery that a human can also operate" tasks: the trend Earthside is already to move humans off of the machinery and remotely operate them (sometimes from a tethered controller nearby, sometimes from a wireless controller nearby, often from a networked controller at an arbitrary distance away). In this situation, sending heavy machinery to Mars with operator cabs so they can be dual-operated by humans or humanoid robots is the short of humorously bass-ackwards thinking that looks cool in hollywood, but has been eclipsed by actual progress: if you want to automate you machinery you automated it by eliminating the physical control step, not putting a robot with robot hands at the controls. Nobody builds self-driving card by putting a robot in the driver's seat of an existing car. Nobody builds UAVs by putting a humanoid robot in the cockpit.

Yes, thank you.

I hope this didn't need to be said, but I'm glad you put it so eloquently.

In general, humanoid robots fill the specific niche of minimising the cost of interacting with existing human-optimised physical infrastructure in existing built environments. When these conditions are not met, robotics optimised for the task have always been the better option - cheaper, more reliable, more efficient.
 
On Mars, there is no bult environment, and there is no existing human-operated equipment. The niche that humanoid robots serve does not exist - and it will only exist if you make design and architecture choices that are actively worse than solutions we already know work well.

This is misunderstanding why humans are on job sites. Yes, some are in operator cabs. But the relevant humans are the ones doing "long tail" tasks: clearing debris out of that finicky tool attachment mechanism, grabbing that dust shield part that came detached so maintenance can bolt it back on, smacking the actuator that gets stuck 0.3% of the time even after repeated actuation cycles.

Wheels are cheaper than legs (especially maintenance), so you just have 1-2 Optimus wrapped up in a "pod" somewhere on the vehicle, suitably dressed in dust-proof bunny suits. When needed they deploy, walk over and do the task, then re-stow on the vehicle.

All the advantages of a vehicle, all the flexibility of an Optimus (while not wasting wear-and-tear trekking kilometers across the desert).

Don't think "operator cab," think "stowed accessory repair bot." Basically R2-D2.

[redneck]

Cross-posting from the humanoid thread because it's relevant.



Don't think "operator cab," think "stowed accessory repair bot." Basically R2-D2.

I'm thinking you've made a case for a repair robot. No necessarily and Optimus or humanoid though. An old joke was that a mechanic should have two elbows on each arm and an eyeball in a finger to work on some of the nearly impossible to reach areas. With robotics, that becomes one of the possibilities. Possibly a Motie Engineer layout with one strong arm with gripping hand and two dexterous arms. Or whatever realistically can work on the various pieces of equipment. I don't see an imperative need for it to have legs or five finger hands. The optimum repair robot may have five arms, a crane, and a forklift for all I know at the moment.

Form follows function is another ancient saying.

[Twark_Main]

Cross-posting from the humanoid thread because it's relevant.



Don't think "operator cab," think "stowed accessory repair bot." Basically R2-D2.

I'm thinking you've made a case for a repair robot. No necessarily and Optimus or humanoid though. An old joke was that a mechanic should have two elbows on each arm and an eyeball in a finger to work on some of the nearly impossible to reach areas. With robotics, that becomes one of the possibilities. Possibly a Motie Engineer layout with one strong arm with gripping hand and two dexterous arms. Or whatever realistically can work on the various pieces of equipment. I don't see an imperative need for it to have legs or five finger hands. The optimum repair robot may have five arms, a crane, and a forklift for all I know at the moment.

Form follows function is another ancient saying.

How cheap/easy/quick do you think it is to develop a human-like capability general-purpose robot? Most people think this technology is a miracle / impossible tier R&D task, and you're casually asking for seconds... 


Unless we can demonstrate (beyond "maybe possibly") that another body-plan is substantially superior and has a shorter path to implement, I expect the first several versions will be the off-the-shelf design already being developed. After all, isn't that the beauty of making them "general purpose" robots?


Obviously simple things like wrist cameras ala Canadarm (or even boroscopes) are easy to add, but that doesn't require developing a new robot.

Eventually we'll live in a world where every kid develops a general purpose robot in science class, and there will be specialized bots for everything. For the next decade, I expect commonality with Optimus will dominate over that sort of body-plan specialization.

[Cheapchips]

Sunday's Memo robot is an interesting wheeled robot, or is it a humanoid with wheels?  The post that it's torso sits on allows it to reach a wider range of positions that you'd expect.  It need smooth floor, but something like this could be useful around a habitat.*  You wouldn't want to move it between floors.  It's bulky and weighs around 45kg.

It's impressive how much dexterity they demonstrate from their grabbers.  They simplify their model training as well.  The teleoperators have 'mittens' which mimic the design on the robot. The input to the model for the actual grabbing and manipulation is 1 to 1.

I assume this is something of what people imagine when they're arguing for two legs bad. 



*when the software matures further, as ever with these things.

[redneck]

How cheap/easy/quick do you think it is to develop a human-like capability general-purpose robot? Most people think this technology is a miracle / impossible tier R&D task, and you're casually asking for seconds... 



Eventually we'll live in a world where every kid develops a general purpose robot in science class, and there will be specialized bots for everything. For the next decade, I expect commonality with Optimus will dominate over that sort of body-plan specialization.

I don't think general purpose will be the target. Craft robot might be a better term for what I think likely. When it's expensive to send mass, special one off designs likely predominate. Not the gold plated flagship nonsense. A shop built unit from COTS parts with a cost commensurate with the mission. I think my main disagreement with some here is in the capabilities of Optimus. I don't see Optimus matching human capabilities in the near future.

[Paul451]

Sunday's Memo robot is an interesting wheeled robot, or is it a humanoid with wheels?  [video]

That is genuinely the most impressive demo I've ever seen. Which is triggering a major "BS alert", but if it's genuine, then the state-of-the-art is much further than anything I've seen from Tesla/etc.

It need smooth floor, but something like this could be useful around a habitat.

While that chassis is low clearance, we know how to build off-road wheeled vehicles. It's not a difficult update compared to... [waves hand]... everything else.

[Twark_Main]

How cheap/easy/quick do you think it is to develop a human-like capability general-purpose robot? Most people think this technology is a miracle / impossible tier R&D task, and you're casually asking for seconds... 



Eventually we'll live in a world where every kid develops a general purpose robot in science class, and there will be specialized bots for everything. For the next decade, I expect commonality with Optimus will dominate over that sort of body-plan specialization.

I don't think general purpose will be the target. Craft robot might be a better term for what I think likely. When it's expensive to send mass, special one off designs likely predominate. Not the gold plated flagship nonsense. A shop built unit from COTS parts with a cost commensurate with the mission. I think my main disagreement with some here is in the capabilities of Optimus. I don't see Optimus matching human capabilities in the near future.

I think we're agreeing past each-other.

On gold plated, I think the opposite. Cattle not pets. You send a pretty large number, so that when a few inevitably break or are accidentally damaged it's no big deal.

On specialist equipment, yes I expect a large majority of equipment mass would be specialized tools and equipment and vehicles.

Optimus weighs 125 lb, with parts and accessories and strap-downs call it 175 lb. So even sending 100 Optimus bots is "only" 8 metric tons of payload mass.   

[Cheapchips]

Fun modular wheeled  / legged / torso bot from LimX.  They fancy themselves as suitable Mars explorer.  Those summersaults shown towards the end would be an interesting way of getting about.

[sanman]

Looks better than that stupid-looking TARS robot from the movie Interstellar