Robot Arm C

[Balsa Man]

Of course this is all unofficial, however we've already started to have some internal committee conversations about this. I strongly encourage someone to submit a question at soinc.org so that we can formally respond there.

So, unofficially, of course, that would mean a solenoid, regardless of what the motion of the plunger was used to do, like to....flip a lever that would say, block, or allow a range of motion of an arm component, would be classified as a motor. Is that what you're thinking here?

[harryk]

A cheap and plausible solution for most teams:
Using two Lego NXT Mindstorms sets, it's possible to create a master-slave system. Although there might be some lag between input and output (I don't think it'll be significant at lower levels). You might end up using 6 motors, but I think it offers a pretty decent solution.

Does anyone have experience with a robot arm that is of similar dimensions to this year's created from Legos? I'm thinking that it might be a little heavy and clumsy though.

I'm not saying this is a bad idea, but unless you already have the kits, the same amount of money could go to making a decent custom arm

[illusionist]

A cheap and plausible solution for most teams:
Using two Lego NXT Mindstorms sets, it's possible to create a master-slave system. Although there might be some lag between input and output (I don't think it'll be significant at lower levels). You might end up using 6 motors, but I think it offers a pretty decent solution.

Does anyone have experience with a robot arm that is of similar dimensions to this year's created from Legos? I'm thinking that it might be a little heavy and clumsy though.

I'm not saying this is a bad idea, but unless you already have the kits, the same amount of money could go to making a decent custom arm

Yeah, our team has like 4 kits combined so... =P

[blue cobra]

Was tipping a problem at Nationals? D batteries probably weigh about 150g and the North Goal is nearly the full 70cm away.

What was successful for moving the parts of the arm? Is a servo the standard? The arm has to be able to fold up into the 30cm square, so I'd imagine that limits things that would extend beyond the arm itself.

What types of control systems are typically used? I have some, but very limited, rc experience (typically just forward/back, left/right controls), and I'd imagine that for each moving part you have you would need another control. What systems are used for these robots, and where could I learn about them?

[harryk]

Was tipping a problem at Nationals? D batteries probably weigh about 150g and the North Goal is nearly the full 70cm away.

What was successful for moving the parts of the arm? Is a servo the standard? The arm has to be able to fold up into the 30cm square, so I'd imagine that limits things that would extend beyond the arm itself.

What types of control systems are typically used? I have some, but very limited, rc experience (typically just forward/back, left/right controls), and I'd imagine that for each moving part you have you would need another control. What systems are used for these robots, and where could I learn about them?

As for tipping over, I'm going with the traditional SO method of bringing a couple 10lb dumbells to weigh down the base
This is a good intro to the mechanics but not so much the electronics:http://www.societyofrobots.com/robot_arm_tutorial.shtml
And this is a torque calculator: http://www.robotshop.com/robot-arm.html(very helpful)

[jander14indoor]

One master slave system showed up at nationals. Worked pretty well, main deficiency was slop in the system. Slave didn't follow master real accurately. That said, it was easy to control and see what was going on. Less slop though would improve time and reduce learning curve.

PS, before counting the servos on the master as motors, see unofficial discussions above.

As to tipping, I didn't see it much, more a problem of arm strength with the batteries and the far goal. I saw more arms buckle or flex excessively than robots tip. As already said, no weight limit, easy to add mass for stability. Smart too.

Arm size vs robot box. Keep in mind its a square, there's no top to that box! The larger devices I've seen have just pointed everything up until the timer starts. I didn't see much in the way of fancy folding.

What worked? Powerful servos. Stepper motors.

Is there a 'standard' solution? Not yet, event is too new.

Jeff Anderson
Livonia, MI

[gh]

By the way, I should note that designing and operating a many-degree-of-freedom (DOF) robot arm is actually quite difficult, from the perspective of computerized control. The more joints you put into the arm, the harder it becomes to solve the problem, "how do I rotate the joints so that the end actuator is at <x, y, z>?"

[ichaelm]

Speaking of computer control, has anyone here done any work on actually using computer programming to control the arm, such as with VEX, Legos, or some other custom system? I would think it could be a big advantage over traditional servo systems, not just because of easier control but also because of the 5-6V limit on traditional hobby servos.

[sj]

Personally i'm against straight computer control of the arm due to the possibility that one slightly not up-to spec piece on the competition board could lead to major disasters. Of course the rules are very explicit about how the board should be set up but even the slightest variation could throw off the whole system. However i believe programing and partial automation is very important. Personally i'm looking into a master slave system once i finish tweaking the actual arm.

[Saturiea]

In theory the best programmed arm would also have some sort of visual sensor to check the board before starting. However this would be a bit unfeasible for most things. Back to reality. I think that a well practiced team can do just as well as a programmed robot in terms of getting the most/all of the points.