pic: 97 revealed ( - end effector + cup of water )

abf7ad6e7ff92eefea86104a3903ac30_l.jpg

With some help from the CD programming forum, we were able to get our gyroscopic arm control code to keep the arm level while the robot is being tilted back (albeit slowly) without spilling any water. :slight_smile:

Details:

Frame: Modified Low-Rider Kit-Bot, 1" ground clearance
Drive: 4 CIM’s with FIRST-supplied gearbox, tank steering
Arm: 5’ tower + 5’ pivoted arm + end effector (still in the works)
Arm Power: Dual FP (12V) w/ gearboxes + 8:1 sprocket reduction
Arm Specs: 177 ft-lb. torque at max efficiency, ~90 degrees/sec max rotation, rate-limited to 30 degrees/sec (for now…)
Autonomous: vision tracking, hopefully, + some backups

WOW! :ahh:

so now you have a waiter bot, huh? I allways wanted one for me :smiley:

Serious:
Have you programmed it by yourselves??

Is it overweight? Guess you’ll need some speed holes…

omg i’m jealous
reminds me of something else i saw here where you called a robot over the intranet and it bought you a coke or something (can’t find it on my search though…)

LOL, It still needs a lot of tuning if it’s going to wait tables. It oscillates quite a bit as is, especially going downward when gravity makes the arm go faster than it wants to. We’re probably going to have to split the code into two cases: one for up and one for down.

I got a lot of help from members of the CD forum, particularly Chris Hibbner and Mike Betts. Their tips helped me figure out why the original code was acting strangely and make it much better. Check this thread: http://www.chiefdelphi.com/forums/showthread.php?t=34673.

As far was we can tell, we are about 15-20 lbs. under without our end-effector, which shouldn’t weigh more than 5 lbs, so we shouldn’t need any speed holes. :wink: