Skip to main content

Inertial Measurement Unit (IMU)

An Inertial Measurement Unit (IMU) is a critical sensor in FRC robotics that measures a robot's orientation, angular velocity, and linear acceleration. IMUs enable autonomous navigation, field-oriented drive systems, and precise robot positioning.

What is an IMU?​

An IMU typically combines several sensors:

  • Gyroscope: Measures angular velocity (rotation rate) around three axes
  • Accelerometer: Measures linear acceleration along three axes
  • Magnetometer: Measures magnetic field strength (compass functionality)

Together, these sensors provide 9 degrees of freedom (9-DOF) of motion data, allowing you to determine your robot's orientation in 3D space. The most important measurements are typically from the gyroscope (heading/yaw) and accelerometer (pitch and roll).

CTRE Pigeon 2.0​

Key Features:

  • 9-axis fusion algorithm for precise heading calculation
  • CAN bus communication for reduced wiring
  • Temperature compensation for improved accuracy
  • Built-in vibration rejection
  • Easy integration with Phoenix framework

Wiring:

CAN High (Yellow) → CAN Bus High
CAN Low (Green)   → CAN Bus Low
V+ (Red)          → 12V
GND (Black)       → Ground

CTRE Docs:

Kauai Labs (Studica Robotics) NavX-MXP​

Key Features:

  • 200Hz update rate for responsive control
  • Multiple connection interfaces (SPI, I2C, USB, Serial)
  • Built-in sensor fusion algorithms
  • Collision detection capabilities
  • Comprehensive software libraries

Wiring (SPI Connection - Recommended): Connect to the MXP port on the roboRIO:

MXP Pin 1  → 3.3V
MXP Pin 2  → DIO 0 (CS)
MXP Pin 4  → DIO 1 (MISO)
MXP Pin 6  → DIO 2 (MOSI)
MXP Pin 8  → DIO 3 (SCLK)
MXP Pin 14 → Ground

Studica Docs:

FRC Applications​

1. Field-Oriented Drive​

IMUs enable field-oriented (field-centric) drive systems where robot movement is relative to the field rather than the robot's orientation.

public void fieldOrientedDrive(double xSpeed, double ySpeed, double rotation) {
    // Get current robot heading
    double robotAngle = Math.toRadians(imu.getHeading());
    
    // Transform driver inputs to field-relative coordinates
    double fieldRelativeXSpeed = xSpeed * Math.cos(robotAngle) + ySpeed * Math.sin(robotAngle);
    double fieldRelativeYSpeed = -xSpeed * Math.sin(robotAngle) + ySpeed * Math.cos(robotAngle);
    
    // Drive the robot
    drivetrain.drive(fieldRelativeXSpeed, fieldRelativeYSpeed, rotation);
}

2. Autonomous Navigation​

IMUs provide critical heading information for autonomous routines:

public class TurnToAngleCommand extends CommandBase {
    private final DriveSubsystem drivetrain;
    private final double targetAngle;
    private final PIDController turnController = new PIDController(0.02, 0, 0.004);
    
    public TurnToAngleCommand(DriveSubsystem drivetrain, double targetAngle) {
        this.drivetrain = drivetrain;
        this.targetAngle = targetAngle;
        addRequirements(drivetrain);
        
        turnController.setTolerance(2.0); // 2 degree tolerance
        turnController.enableContinuousInput(-180, 180);
    }
    
    @Override
    public void execute() {
        double currentAngle = drivetrain.getHeading();
        double turnSpeed = turnController.calculate(currentAngle, targetAngle);
        drivetrain.arcadeDrive(0, turnSpeed);
    }
    
    @Override
    public boolean isFinished() {
        return turnController.atSetpoint();
    }
}

3. Balance Detection​

Use IMU pitch readings to detect when your robot is on the charging station (2023):

public boolean isOnChargingStation() {
    double pitch = Math.abs(imu.getPitch());
    return pitch > 10.0; // Robot is tilted more than 10 degrees
}

public boolean isBalanced() {
    double pitch = Math.abs(imu.getPitch());
    return pitch < 2.5; // Robot is within 2.5 degrees of level
}

Troubleshooting​

  1. IMU not detected: Check wiring and CAN/SPI connections
  2. Inconsistent readings: Ensure proper mounting and calibration
  3. Drift over time: Implement periodic resets and consider environmental factors
  4. Noisy data: Add software filtering or check for mechanical vibration