Robot Oriented vs. field Oriented

[BEN35678]

Do any teams have experience using field orientation style of driving and if so please give some pros and cons of using this style.

[Iaquinto.Joe]

We're using robot centric mecanum drive for now. The mecanum cartesian algorithm allows you to switch pretty easily so for us it's just driver preference. I like field centric better myself because it reduces complexity for the driver. I always explain it like you're sitting on the robot and holding onto the controls. Push left, go left. Push forwards, go forwards. That's robot centric. Field centric is when you're behind the glass and your robot is facing due east. Push forwards, robot moves away. Push left, goes left according to your frame of reference. One less thing for the driver to handle if they don't have to project their frame of reference onto the bot. Robot centric is probably more reliable over time though.

[Ether]

Robot centric is probably more reliable over time though.

Why? Because of gyro drift? Or did you have something else in mind?

[Iaquinto.Joe]

Why? Because of gyro drift? Or did you have something else in mind?




Yes gyro drift. We've looked into several different IMUs for an alternative.

[Ether]

Yes gyro drift.

One thing teams have done to mitigate this is to have a button on the joystick that the driver can press to zero the gyro whenever the bot is in the zero position (usually facing straight downfield).

[Pratik Kunapuli]

One thing teams have done to mitigate this is to have a button on the joystick that the driver can press to zero the gyro whenever the bot is in the zero position (usually facing straight downfield).




Something that our team does is to click the B button on the driver's gamepad before the match starts but after the robot is placed on the field, and this zeros the gyro. We usually try to do this as late as possible, but it is very effective in eliminating drift. Something else that you could do is to turn on the robot after it is placed in its starting configuration so that the gyro doesn't change when teams position the robot.

[Iaquinto.Joe]

Something that our team does is to click the B button on the driver's gamepad before the match starts but after the robot is placed on the field, and this zeros the gyro. We usually try to do this as late as possible, but it is very effective in eliminating drift. Something else that you could do is to turn on the robot after it is placed in its starting configuration so that the gyro doesn't change when teams position the robot.

We're planning on implementing a running average of our voltage the first time we're in disabled (On startup) of something like 32 samples. When we get enabled, our code would have our calibrated voltage of when the match countdown is going on.

[jdunston94]

I drove for my team senior year of HS, and a little bit my Junior year. I have driven both and i prefer robot oriented way above field orientation but that's just the way i think. it really is driver preference, i have seen teams with a toggle option to say whether they wanted field or robot orientation mid match though i think that would get a little crazy.

[Matt C]

Team 190 pulled off a rather successful field oriented control using custom mechanum wheels back in 2005. Back then though we were able to make a segmented frame that had a pivot so that each wheel would remain in contact with the floor. From what I remember, the system also brought the old IFI control system to it's knees.

The coolest thing that they were able to do with a field oriented control was spin the robot around while moving in a straight line.

[connor.worley]

1717 has done some cool stuff to get field-centric swerve. I would look to them.