Multi-IMU Fusion
Modular Python package for sensor fusion across multiple IMU types
- Status
- Research
- Version
- 1.0.0
- Platform
- Library / CLI
- License
- MIT
Combining orientation data from different IMU models usually needs per-device drivers, calibration and fusion code; this package puts them behind one API.
How it is put together.
Multi-IMU Fusion is a Python package for reading and fusing data from several IMUs at once. It includes drivers and auto-detection for common IMUs over I2C/SPI/UART, gyro and accelerometer calibration, Madgwick, EKF, UKF and complementary filters, and multi-rate synchronisation. Fused data can be logged to CSV/HDF5 or streamed over UDP, TCP, WebSocket or MQTT, and a PyQt5 GUI and CLI provide control and 2D/3D orientation views.
- 1
IMU drivers
Hardware abstraction per IMU and auto-detection
multi_imu_fusion/imu/ - 2
Fusion core
MultiIMUFusion: manages IMUs, runs fusion, exposes API
multi_imu_fusion/core.py - 3
Fusion algorithms
Madgwick, EKF, UKF, complementary
multi_imu_fusion/fusion/ - 4
Calibration
Gyro and accelerometer calibration
multi_imu_fusion/calibration/ - 5
Streaming and logging
Network streaming and file logging
multi_imu_fusion/data/ - 6
GUI and CLI
PyQt5 GUI and command-line interface
multi_imu_fusion/gui/
Specification.
- IMU drivers
- MPU6050, BNO055, LSM9DS1, ICM20948, Xsens; plugin interface for new devices
- Buses
- I2C, SPI, UART with auto-detection
- Fusion
- Madgwick, EKF, UKF, complementary filter
- Calibration
- Gyroscope and accelerometer calibration routines
- Streaming
- UDP, TCP, WebSocket, MQTT
- Logging
- CSV and HDF5
- Interfaces
- Python API, CLI (status, list, calibrate, stream, log), PyQt5 GUI
- Benchmarks
- Committed benchmark result JSON files (benchmark_results/, 2025-07-06)
- Runs on
- Python package (pyproject.toml, name multi-imu-fusion) with CLI and optional PyQt5 GUI; needs IMU hardware
More systems.
Turn any Android phone into a georeferenced mobile-mapping and Visual-Inertial SLAM rig.
Mapping and localization · ResearchCORRIDYXTurn dash phones into a live, city-scale map of where camera-based autonomy can see
Next systemnumberplate-detectionReal-time vehicle and license plate detection and recognition