IMU Calibration System

Python-based calibration system for IMU sensors (accelerometer, gyroscope, magnetometer), following the PT calibration pattern.

Files

• imu_calibration.py - Core calibration library
• imu_calibration_gui.py - Interactive GUI for calibration
• accelerometer_calibration.py - Accelerometer-specific calibration script
• gyroscope_calibration.py - Gyroscope-specific calibration script
• magnetometer_calibration.py - Magnetometer-specific calibration script

Usage

Accelerometer Calibration

Static Position Method:

python accelerometer_calibration.py --interactive --output accel_cal.json

Place sensor in 6+ orientations where gravity is the only acceleration:

• +X, -X, +Y, -Y, +Z, -Z axes up

Gyroscope Calibration

Zero-Velocity Method:

python gyroscope_calibration.py --duration 60 --output gyro_cal.json

Keep sensor stationary for specified duration. Calibration estimates bias.

Magnetometer Calibration

Ellipsoid Fitting Method:

python magnetometer_calibration.py --interactive --output mag_cal.json

Rotate sensor through 12+ orientations. Calibration estimates hard iron (bias) and soft iron (scale/misalignment).

GUI Mode

python imu_calibration_gui.py --sensor-type accel --port 5008

Interactive GUI for real-time calibration with live plots.

Calibration Parameters

Each calibration produces:

• Bias: Offset correction [sensor units]
• Scale Matrix: Scale factors and misalignment [3x3]
• Temperature Coefficient: Temperature compensation
• Quality Metric: Calibration quality (0-1)

Integration with FSW

Calibration parameters can be loaded in C++ FSW:

#include "calibration/IMUCalibration.hpp"
 
// Load calibration from JSON
auto calibrator = std::make_shared<IMUCalibration>(SensorType::ACCELEROMETER, "imu_0");
// Would need JSON loader in C++

Or use Python calibration results directly in Python-based systems.