[Dprglist] Iron Reign presentation

[John Swindle]

I enjoyed the presentation by Iron Reign at the last DPRG meeting. Great team, great design.
As an armchair designer, I have a suggestion which might be able to be used before competition:
The team was going to change the mechanical damping (the squeeze on the squishy plastic hinge) to reduce the time that the gripper wobbles. That is probably sufficient, but here's something that might make the entire platform behave as if it were on rails:
PWM the robot's traction, turntable, and arm movements so that movements are critically damped (or nearly so) in movement profiles. I've seen the larger 120 pound robots extend to 6 feet where they were top-heavy and yet start and stop on a dime with absolutely no wobble (and I've seen other 120 pound robots doing the same thing and wobbling a lot). I know the critical damping was accomplished by determining the responses of the platform and programming those responses into the movement profiles. It was not based on sheer mass in the bottom of the platform.
To be fair, the Bode plots for that nice robot's movement profiles were done by the mentors and not by the team, and I am impressed how much the Iron Reign team accomplishes on their own independently of sponsor or mentor input.

I was impressed with the idea of using a spline for high-speed movement between two points in the arena. I use splines in my audio processing because they magically mimic the responses of real mechanical systems with no prior knowledge about the system, other than the measured points. I had not thought of using a spline for XY movement of a robot. Neat idea.
And, actually, when using a spline, the movement profile of the robot needs to be fairly accurate, else the spline is not realistic.
Good stuff!
And I loved Carl's tour of the Pittsburgh International Airport Branch of the Robot Repair store.
John Swindle
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.dprg.org/pipermail/dprglist-dprg.org/attachments/20200211/1f7db0cc/attachment.html>