Idea of a pitch marking robot

hello sir;
I am working on a project of a pitch(Ground) marking robot which marks the boundaries of different sports ground. I just want to know whether can we use a GPS system for this?? If yes than what could be the accuracy required ?? if not please suggest some ideas about it...???
Please help me ..... I an a beginner in robotics.

Thanking you in advance

Ordinary GPS is +- 1m, compensated GPS +- 10 Cms. what is the tolerance on pitch dimensions?
Frank

I'm also a beginner in robots and I have thought about a similar application. These are my thoughts:

A GPS can be used although it may not be a requirement. I believe that in order to obtain reasonable accuracy (so your markings don't "zig-zag") encoders on wheels will be required so that you know exactly how far your 'bot has traveled. On the left and right wheels so you know that each wheel has rotated the expected amount.

If they are onto drive wheel, you run into problem if the wheels slip or spin on a slippery surface. So maybe an extra set of non-drive wheels with encoder which simply roll on the ground will give accurate travel information if the drive wheels slip.

With encoders, you could simply program the motors to travel exact amount w/o needing needing a GPS or compass because you can infer the 'bots position by the encoders alone.

With a GPS and compass, you simply "punch in" the necessary GPS coordinates and program your code Use the GPS and compass data to orientate your 'bot in the correct direction, and calculate the "point a to point b" vector with the coordinates you programmed in, and with that, know the number of rotations the wheels required to travel the distance.

Have fun and good luck

hello sir;
We have to compulsorly use a different type of mechanism called as Theo jansen's Mechanism for building the chasis of robot instead of normal wheel robots. So please give any idea assosciated with it. Can we use only GPS Module and Compass for it??? What about the accuracy of pitch marking by using this two systems???
Please reply..
Thank you in advance

After some expectable accuracy numbers have been mentioned, it would be fair to tell your requirements.

I expect that said DGPS accuracy of 10 cm is not sufficient, and adding a compass sensor can't improve the accuracy.

So What i could use with it to improve the accuracy of the pitch marking.??? Should i change over from gps to some other technique or GPS of accuracy +-10 cm would be ok.......Please reply...
Thank you in advance

Read about DGPS

FvM said:

Read about DGPS

Click to expand...

As for this idea I need highest accuracy possible
could image processing be used for this idea as an alternative option for bigher accuracy than GPS ??? Can u suggest some method having higher accuracy..??????

Theo jansen's walker might put in an angular deviation on a given step due to the ground not being absolutely flat, and I can't see how it would recover. I would use a tracked vehicle. I think I would put down a base line of three ultrasonic radiators in a straight line and equal distance apart. Send a timed pulse to each one in turn then it should be able to deduce where your robot is with respect to the base line. i.e. if Time b1-b2 = time b2-b3 then the robot is on axis. Even this is fraught as the speed of sound varies with the air temperature and pressure. .
Frank

Don't forget that an ultrasonic measurement in open field is also a wind speed meter.

Laser time-of-flight or triangulation would be more reliable.