View Single Post
  #28   Spotlight this post!  
Unread 07-12-2010, 01:35
AustinSchuh AustinSchuh is offline
Registered User
FRC #0971 (Spartan Robotics) #254 (The Cheesy Poofs)
Team Role: Engineer
 
Join Date: Feb 2005
Rookie Year: 1999
Location: Los Altos, CA
Posts: 803
AustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond reputeAustinSchuh has a reputation beyond repute
Re: Definitive FRC Mechanisms

Quote:
Originally Posted by Ether View Post
I'm trying to parse the above. What I think I hear you saying is this:

Since backing up the robot tended to rip the ball out of the roller mechanism, you realized you needed more torque on the roller when backing up (and perhaps less torque when going forward?). So instead of using a constant setpoint for the current command, you subtracted some fraction of the drivetrain command from the roller current setpoint. Clever. Am I understanding it correctly ?
Yes, you are. That's definitely a bit clearer than what I originally wrote. We originally did this by adding some more to the voltage that the current loop on the cRIO was sending to the jaguar.

I'm not totally sure if the real issue is that you need more torque when backing up, but that the ball would roll on the ground when backing up. This would require us to spin the roller up, but still have the same torque applied to the ball from the roller. The acceleration of the roller due to the current loop wasn't fast enough to react to this correctly, so our little bit of extra umph would help that out.