1D Motion Estimation with Kalman Filter

This repository contains a Simulink-based implementation of a Kalman Filter for estimating the position and velocity of a vehicle moving in one dimension using noisy sensor measurements.

Project Overview

The system models a vehicle receiving torque inputs and simulates its one-dimensional motion. A Kalman Filter is used to estimate the state vector:

The goal is to reconstruct accurate state estimates using:

• Noisy position measurements (GPS-like)
• Known system dynamics based on torque input and mass

Features

• Full implementation of a discrete Kalman Filter in Simulink
• Separation of prediction and update steps into distinct Simulink subsystems

Included Files

• kalman_filter_data.mat: contains all simulation data and parameters
• Simulink model implementing the Kalman Filter
• MATLAB scripts for loading data
• Output graphs for result visualization

How to Run

1. Open MATLAB and load the data.
2. Open and run the Simulink model (kalman_filter.slx).

Output Plots

Technical Notes

• Sample time: dt (loaded from .mat file)
• Solver: Fixed-step (step size = dt)
• Compatible with MATLAB R2021a