This project simulates a signal processing pipeline for noisy sensor data using a Kalman Filter, then feeds the filtered estimates into a feedback controller that drives a simulated actuator. The goal is to demonstrate how estimation techniques improve system performance in closed-loop control applications.
- Simulates noisy 1D position measurements (sensor readings).
- Applies a Kalman Filter to estimate true position and velocity.
- Implements a simple proportional controller to generate control commands.
- Simulates actuator response over time and visualizes tracking performance.
- Shows how filtering enhances actuator stability and accuracy.
kalman-filter-sensor-actuator-simulation/ ├── Kalman_Filter_Sensor_Simulation.ipynb # Main simulation notebook ├── kalman_actuator_tracking.png # Example output plot (optional) ├── README.md # Project documentation └── .gitignore # Ignore virtualenv, notebooks checkpoints, etc.
- Python 3.x
- Jupyter Notebook
- Libraries:
numpy,matplotlib,filterpy
The actuator uses the Kalman-filtered position to generate control actions and track the desired trajectory. The filtered estimates significantly improve stability compared to raw sensor readings.
-
Clone the repository:
git clone https://github.com/Arman-Rajaei/kalman-filter-sensor-actuator-simulation.git -
Install dependencies:
pip install numpy matplotlib filterpy -
Open
Kalman_Filter_Sensor_Simulation.ipynbin Jupyter Notebook. -
Run all cells to simulate the system and generate output.
This project demonstrates practical state estimation and control system design using a Kalman filter and simulated actuator feedback. It aligns well with roles focused on:
- Control engineering
- Signal processing
- Robotics and automation
- Space or embedded systems engineering
Arman Rajaei