Guidance, Navigation and Control - Engineering Portfolio

Three from-scratch GNC projects with deterministic, reproducible validation, built as a portfolio for aerospace, mechatronics, controls, and test-engineering roles.

Guidance says where to go. Navigation says where you are. Control closes the loop.

What's Inside

#	Project	What it does	One-line pitch
1	projects/attitude_control	3-axis spacecraft attitude simulator with PID and LQR-style feedback	Rigid-body dynamics plus closed-loop attitude control for pointing
2	projects/kalman_filter	Extended Kalman Filter for noisy IMU and GPS sensor fusion	Real-time state estimation from messy sensors
3	projects/embedded_controls	Arduino closed-loop PID plus desktop plant simulator	Hardware-side controller with a software twin for safer tuning

Each project folder has its own README with theory, equations, usage, tuning notes, and resume talking points.

Results Snapshot

All numbers below are produced deterministically by scripts/run_all.py and checked by scripts/validate.py.

Project	Metric	Result
Attitude Control	Final attitude error	0.5032 deg
Attitude Control	Final body-rate norm	0.00361 rad/s
EKF Sensor Fusion	Position RMSE	0.6911 m
Embedded PID	Final tracking error	0.0494 units
Embedded PID	Overshoot	2.9259 units

Visual Results

Attitude Control	EKF Sensor Fusion	Embedded PID
Attitude error vs time for the demo slew	EKF estimate overlaid on the truth trajectory	Desktop plant response under closed-loop PID

Quick Start

Linux / macOS

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python -m pytest -q
python scripts/run_all.py

Windows PowerShell

python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
python -m pytest -q
python scripts\run_all.py

Generated CSV files and plots are written to results/.

If you do not want to install pytest, the built-in validation checks still run:

python scripts/validate.py

Skills Demonstrated

• Control systems - PID tuning, LQR-style state feedback, overshoot tracking, settling behavior, actuator limits
• State estimation - Extended Kalman Filter, IMU/GPS sensor fusion, noise modeling, gyro-bias estimation
• Rigid-body dynamics - quaternion kinematics, body rates, Euler rotational equations, RK4 integration
• Embedded control - Arduino sensor-read to actuator-write loop, serial gain tuning, PWM output, anti-windup
• Verification engineering - deterministic seeds, validation thresholds, pytest tests, GitHub Actions CI
• Python toolchain - NumPy, Matplotlib, pytest, reproducible scripts

Expected Validation Thresholds

python scripts/run_all.py should produce:

• Attitude error below 1.0 deg
• Body-rate norm below 0.02 rad/s
• EKF position RMSE below 1.5 m
• Embedded PID final tracking error below 1.0 unit

Small numerical differences are normal across Python and NumPy versions.

Project Layout

projects/
  attitude_control/      # 3-axis rigid-body dynamics + PID / LQR-style feedback
  kalman_filter/         # EKF for IMU + GPS sensor fusion
  embedded_controls/     # Arduino firmware + Python plant simulator
scripts/
  run_all.py             # Runs every demo end-to-end
  validate.py            # Deterministic threshold checks
tests/                   # pytest suite
docs/images/             # Generated result plots
.github/workflows/       # GitHub Actions CI

Author

Prajwal Bekal
M.Sc. Mechatronics and Cyber-Physical Systems, Deggendorf Institute of Technology
GitHub | LinkedIn