Unscented Kalman Filter

Unscented Kalman filter using LiDAR and Radar feed.

Overview

This project is part of Udacity's Self-Driving Car Nanodegree program. An Unscented Kalman Filter (UKF) has been implemented in this project, where LiDAR and Radar measurements are fused to predict the position and velocity of a simulated car. Constant Turn Rate and Velocity Magnitude model is used for the state-vector, which contains 2-D position coordinates, velocity, yaw-angle and yaw-rate of the observed object as its components.

A simulator provided by Udacity is used to generate and visualise measurements and motion of a car. More information on installation and usage of the simulator with the executable can be found in the seed-project setup by Udacity here.

Dependencies

1. CMake >= 3.5
2. Make >= 4.1
3. Eigen 3.3.5
4. gcc/g++ >= 4.1

Build and Run Instructions

1. Create a build directory in the parent directory

mkdir build

2. Run CMake and make in the build/ directory

cd build; cmake ../; make

3. Launch the simulator
4. Run the UKF executable

./UnscentedKF

Notes

1. For the given sensor measurements provided by the simulator, RMSE errors in the prediction of the car's state (position and velocity) were observed as follows:

Dataset Index	RMSE Position x	RMSE Position y	RMSE Velocity x	RMSE Velocity y
1	0.0693	0.0835	0.3336	0.238
2	0.0685	0.0693	0.5846	0.2473