Although our season is over, the programming team is still tasked with completing several things before the start of the next season. One of the more difficult things (and something I struggled with implementing) during the build season was using a gyro to help the drivers. This is the code I’ve come up with so far, but I only included the drive code, nothing else:
If your goal is to use the gyro to provide a field-centric driver interface, that’s already part of the Java WPILib:
*
* /**
* Drive method for Mecanum wheeled robots.
*
* A method for driving with Mecanum wheeled robots. There are 4 wheels
* on the robot, arranged so that the front and back wheels are toed in 45 degrees.
* When looking at the wheels from the top, the roller axles should form an X across the robot.
*
* This is designed to be directly driven by joystick axes.
*
* @param x The speed that the robot should drive in the X direction. -1.0..1.0]
* @param y The speed that the robot should drive in the Y direction.
* This input is inverted to match the forward == -1.0 that joysticks produce. -1.0..1.0]
* @param rotation The rate of rotation for the robot that is completely independent of
* the translation. -1.0..1.0]
* @param gyroAngle The current angle reading from the gyro. Use this to implement field-oriented controls.
*/
public void mecanumDrive_Cartesian(double x, double y, double rotation, double gyroAngle) {
if(!kMecanumCartesian_Reported) {
UsageReporting.report(tResourceType.kResourceType_RobotDrive, getNumMotors(), tInstances.kRobotDrive_MecanumCartesian);
kMecanumCartesian_Reported = true;
}
… Or if your goal is to make the bot drive straight when the driver is commanding zero rotation, you could do something like this:
…Or here’s an interesting optional (button-press selectable) driver interface mode for mec that your drivers might like:
Halo Auto-Rotate (Halo-AR) Driver interface
A single 2-axis joystick does field-centric control of forward/reverse
and strafe right/left motion, while the robot automatically and
simultaneously rotates to face the direction it is commanded.
Wait so I can just do
RobotMap.robotDrive.mecanumDrive_Cartesian(OI.joystick0.getX(), OI.joystick0.getY(), OI.joystick0.getZ(), RobotMap.driveGyro.getAngle());
in my teleopPeriodic and it’ll do corrections for me? So if I’m driving w/o rotating it’ll auto correct for me?
I think I’m confused about what field-centric really is. To my understanding it’s positioning based on the layout of the field. Could someone explain this to me?
Oh I see. That is not what I’m trying to accomplish. I am trying to get the robot not to rotate when the driver rotate command is zero. How do you accomplish this in Java, not in Labview?
I recommend using one of the WPIlib MecanumDrive methods, and just make corrections to the “twist” input when you want to go straight based on the error. Do not use the fourth “gyroscope” input to MecanumDrive_cartesian; this is only used for “field coordinates”.
The correction could be a full PID, but because of the amount of friction in a mecanum drive system, you should be able to get away with a simple proportional correction; D will be mechanical, and I is usually not needed when seeking a position.
To do this, you will need to create a numeric variable to hold the desired heading, say azi.
In your joystick-reading loop:
When the joystick values indicate that you do not wish to hold a constant facing, set azi to NaN, or some number outside of the valid range.
When the joystick values indicate that you **do **
wish to hold a constant facing:
[LIST]
If isnan(azi), set azi =gyro.getAngle()
otherwise, add a correction of the form prop * (azi - gyro.getAngle()) to the “twist” value. You will have to tune the value of prop.
[/LIST]
Debugging tip: If the robot starts spinning faster and faster, do a quick shut down, change the sign of “prop” and try again.
mecanumDrive_cartesian() is a method within the standard RobotDrive class in both C++ and java WPIlibs. For what are hopefully obvious reasons, you must use one of the four-controller forms of the new RobotDrive() constructor in order to use the mecanum methods.
As explained in earlier posts in this thread, passing the gyro angle to the fourth parameter gives you field-centric control. You said that’s not what you want.
See the note “Optional Input for field-centric control of XY motion” in the code diagram linked in this post:
The code is not particularly well documented, but here’s the code we used this year:
We pass in the rotation joystick value that has already been deadband’ed.
If the joystick is at zero rotation, we use a proportional controller to keep the robot straight.
If the joystick is not at zero, we update the current heading of the robot and pass the joystick value back to the drive system unchanged.
We also have the option to disable gyro mode at any time in case something goes wrong like a bad calibration, or a loose cable.
There is also an option to update the kP value from the smart dashboard and some diagnostic information sent back to help troubleshoot when things go wrong.