Udacity Self-Driving Car Engineer Nanodegree Program, Term 2 ,Project 2 - Unscented Kalman Filter
This project is write by C++.
I used Unscented Kalman Filter to predict and update the position based on lidar and radar data.
This project involves the Term 2 Simulator.
I was build in Ubuntu 16.04:
- Clone this repo.
- Make a build directory:
mkdir build && cd build
- Compile:
cmake .. && make
- Run it:
./ExtendedKF
There are two dataset.
px:x-position
py:y-position
vx:velocity in the x-direction
vy:velocity in the y-direction
MSE:mean squared error
The UKF accuracy was:
Dataset 1 : RMSE = [0.0724, 0.0822, 0.3425, 0.2304]
Dataset 2 : RMSE = [0.0882, 0.0713, 0.6632, 0.3073]