Mobile Robot System

ExcessiveO wrote:

Is there a reason to not just put the encoders on the driven wheels? I'd be skeptical of relying on encoder wheels for positioning. There's a high risk of it slowly drifting off course due to dust/slipping of the encoder wheels. meaning when it travels down and back, the cuts will almost definitely not line up at the end. Some absolute positioning system would be best.
 

What happens with the encoders mounted on the driven wheels is that they are not able to account for the slippage between the wheels and the sheet metal. However, with the externally spring loaded pivoted mounting, it specifically tries to counter this problem.

Also, cable transducers are very expensive for such long lengths (work area required is 15m x 3m). And the error is substantial over longer distances according to a manufacturer. 
And vision systems seem impractical in a plasma cutting environment.

Grotius wrote:

If you have 2 of above devices. And a digital compass. I think it will work.
You have to unpack the devices and use the lenght input as input for x and y. Then the compass will tell the robot if the rotation value is correct.

I had thought about using two such laser distance measurement devices with reflective surfaces at the boundary of the work area so the readings will be further more accurate. There are two problems though. In continuous measurement mode, the measurements are made every half second on an average. So, the frequency seems insufficient. And the second part is interfacing the data into the system. Can we read the distance data into our system from their proprietary devices? 

arvidb wrote:

I think your best bet is to make your robot move in only one axis, and make it 3 meters wide - i.e. build a gantry on wheels!
 

Good thought! It will be my last resort.

How could we go about incorporating dual position feedback?
1. Closed loop steppers or Open loop steppers with external encoders
2. Laser or some other Distance Measurement Sensors
How would we go about with PID control with two feedbacks?
Is dual feedback even a good idea?
One line of thought would be splitting the positioning system into local (encoders) and global (laser)

What are all the other constraints for the system? Cost, max size including Z, how soon do you need/want it, max weight, how fast does it NEED to cut, how much can you do, who can help you, etc etc

So far I see:
Plasma cutter - mobile on its own wheels
15m x 3m  XY
1mm accuracy of cut

@dhruv  I beg to differ. If fiducials are placed on the material as well as the robot a camera system can easily keep track of the cutting head from above. Some big fans are cheap and can provide cross flow over the cutting area to keep the sightline clear. I don't know what his max Z is allowed to be.

2 HD cameras placed above per sq meter will easily get the 1mm accuracy.

Bari wrote:

So far I see:
Plasma cutter - mobile on its own wheels
15m x 3m  XY
1mm accuracy of cut
 

I’m based in India and this is my Under Graduate level project and I have three other team mates. We have been assigned two project guides from my college. The project is funded by a consultancy and the initial budget is INR 1 lakh. I want to complete it in the next 3-4 months. I have a little experience with mechatronics and robotics however this is my first project with LinuxCNC.Now, the sheet metal is placed on the ground on spacers (They don’t use a water bed) and the mobile bot will traverse on the surface of the sheet metal. The bot has a 4 Omni Wheel drive with X (&U), Y (&V), Z (15cm travel) and A axis (for bevel cutting) motion.
 
If the budget is overshooting by a lot, then I may keep the bevel cutting rotary DOF to be manual. The cutting speed is 1m/min. There is no weight limit as such.

@Grotius about the laser distance measurer from Bosch, I talked to a representative for laser distance measurer and he said that interfacing can't be done with their products. However, on a forum for sketchup, some users were discussing a similar topic and they were of the opinion that such interfacing is possible in Leica products. However, traditional laser distance sensors seem like a better option. I found some appropriate sensors from micro epsilon, pepperl-fuchs and banner engineering and will enquire accordingly.