⚠️ The repositories of this project have been transferred to this organization from @ShivtejG236.

Repo 1 • 500 Hz deterministic multi-sensor data acquisition firmware for real-time navigation systems that fuses data streams from IMU, magnetometer, barometer and GPS into a timestamped CSV over UART at 921600 baud.

Repo 2 • Deterministic serial logging of the sensor data, automated session trimming, robust sensor calibration pipeline and graphs & timing reports generation (for validation purposes only).

Repo 3 • Downstream processing of the calibrated CSVs using Extended Kalman Filters & Allan variance for map visualization & diagnostic reports. Focuses on sensor error modelling, state estimation accuracy and filter convergence.

Repo 4 • GPS dropout modelling, sensor fault injection, latency benchmarking, downsampling analysis and (later) TinyML quantization experiment & motion simulation in near-real-time using Nvidia's PhysicsNeMo model for AI surrogate generation.

Repo 1	Repo 2
task_sensors = producer task_serial = consumer

Hardware Setup Component Interface Address IMU I²C 0x68 Magnetometer I²C 0x0D Barometer I²C 0x76 GPS UART2 9600 baud Host logger UART0 921600 baud
Sample Rates Sensor Rate Period MPU-6050 500 Hz 2 000 µs QMC5883L 75 Hz 13 333 µs BMP280 25 Hz 40 000 µs NEO-6M 1 Hz 1 000 000 µs
All three I²C devices share the same bus at 400 kHz fast-mode Lower-rate sensors are polled sub-sampled within the 500 Hz IMU loop — no separate timers, no RTOS overhead

firmware/
├── src/
│   └── main.cpp
├── include/
│   ├── config.h
│   └── sensor_row.h
├── platformio.ini
├── sdkconfig.defaults
└── docs/
    └── setup.png
            

data-pipeline/
├── data/
├── outputs/
│   ├── plots/
│   └── reports/
├── pipeline/
│   ├── logger.py
│   ├── session_trimmer.py
│   ├── diagnostics.py
│   ├── calibrator.py
│   └── plotter.py
├── run_pipeline.py
└── README.md
            

# Build
                pio run
                            

# Flash + open monitor
                pio run --target upload && pio device monitor --baud 921600
                			

# Environment
                python -m venv .nav
                source .nav/bin/activate
                pip install -r requirements.txt
                			

# Port
                /dev/cu.usbserial-XXXX on macOS (usbserial-0001)
                /dev/ttyUSBX on Linux (ttyUSB0)
                COMX on Windows (usually COM3 with no other UART devices connected)
                			

# Execution
                python run_pipeline.py	# Interactive mode
                python run_pipeline.py --action x --session-dir data/session_XYZ --calib-mode y	# Automated mode
                			

Repo Status firmware ✅ Active data-pipeline ✅ Active analysis 🚧 In progress edgehard 🔜 Planned