Should remotely operated vehicles be on topic on Robotics?

Question

The recent meta discussion about the auv, uav, ugv and (possible) drone tag taxonomy, and my subsequent suggested tag wiki edits (still pending as I write this) got me thinking about whether we should consider remotely control vehicles on topic on Robotics.

My initial reaction was that no, ROVs probably shouldn't be on topic, and should probably be referred to Electrical Engineering stack exchange - so my tag wikis reflect that. On second thoughts though, I'm not so sure and I think it is worth discussing.

---

During definition, we only had a single example tele-presence question suggested (as far as I can see) and it received no votes either way, so that doesn't help.

Considering the similarities, I can see that many of the control elements of autonomous and remotely operated vehicles are similar, if not the same and many of the engineering challenges for autonomous mobile robots are identical to that of remote controlled mobile robots.

Having said that, I was always disappointed that Robot Wars was as much, if not more, about the driver of each robot than it was the robot itself. I wanted to see real robots battling it out using their own sensors and their own autonomous decision making processes to win the day.

Then there are the grey areas, remotely operated vehicles which take their higher level control from a human operator, but do significant on-board processing to make them stable and hold position when not being actively controlled. These feel to me like they should be on topic, as they could easily be a stepping stone to or a component of an autonomous system.

Answer 1

IMO, ROVs are very relevant to robotics, just slightly easier as they don't have complex programming involved. There still are intricate mechanical aspects of the thing, and generally you still program them to some extent (for example, to make differential turning smooth). Recently, I've been working on what amounts to an ROV (for ASME SDC 2013) -- and from this perspective, yes, an ROV is robotics.

Though I sense you're talking about questions pertainint to the remote control part. Then yes, EE.SE is better (though generally we use readymade wireless circuits like Zigbees). If it's about those readymade circuits, it seems OK here, but if it's about making/modifying your own RF circuit, EE would be the place to go.

Answer 2

IMO, developing robots usually involves two main parts, building the hardware (engineering) and programming the software (computer science). Computer scientists tend to involve themselves in the autonomy end of things (think Roomba and the Google car), because that's their forte. Engineers, on the other hand, are usually more interested in building ever more impressive hardware (think BigDog and Cheetah) without worrying much about autonomy.

So I think autonomy is a sufficient condition for a machine being classified as a robot, but not necessary. Or to put it more succinctly: "If it looks like a duck, swims like a duck, and quacks like a duck, then it's probably a duck."

So far, our policy has been that if a question is related to robotics, we'll allow it even if other SE sites could probably give better answers. I think this is a sane way to go. On a day-to-day basis, I'd recommend that if a question is borderline and doesn't garner enough interest or successful answers within a reasonable time frame, we consider moving it to a more topical SE site. But, in general, I'd err on the side of inclusion.

Answer 3

Do I command it to go to the next room, or to drive forward? That is where I draw the line. Why would a RC plane be considered a robot? Why would a pan and tilt camera be considered a robot?

If I command it to follow a line, then it barely is a robot. If I command it to hold a GPS position, it also is barely a robot, And in most case a Quad UAV potentially fits this description.

But a telepresence robot - no, that's EE. Note that even JPL classifies their equipment not as robots, but as robotic rovers. And DaVinci Surgical System uses four interactive robotic arms, but it is NOT a robot.

Personally I would very much like to see the term robotics, not be mixed with mechanical and electrical engineering. I understand they are a part of, but not the same as robotics.