CarND-Unscented-Kalman-Filter

Self-Driving Car Engineer Nanodegree Program

---

Description

• This is the 2nd project in term 2 of Udacity self-driving cars nano-degree. It implements UKF in C++ to track a bicycle given lidar and radar sensors
• UKF is an alternative technique to normal or extended KFs. It deals with non-linear process and measurement models … instead of linearizing non-linear equations, it uses sigma points to approximate probability distribution.
• Advantages: - Better approximation of non-linear motion - More efficient (no Jacobian matrix calculation)

Dependencies

• cmake >= 3.5
• Used installer: cmake-3.7.2-win64-x64.msi
• make >= 4.1
  • Used installer: make-3.81.exe
• gcc/g++ >= 5.4
  • Used installer: mingw-get-setup.exe

Basic Build Instructions

Once you have this repository on your machine, cd into the repository's root directory and run the following commands from the command line:

mkdir build && cd build
cmake .. && make
UnscentedKF (path_to_input).txt (path_to_output).txt
    - eg. `UnscentedKF ../data/obj_pose-laser-radar-synthetic-input.txt output.txt`

NOTE

If you encounter any problems, copy "vcvars32.bat" to build directory and run the command vcvars32 to set environment variables

If make command does not work try: cmake .. -G "Unix Makefiles" && make

You can find some sample inputs in 'data/'.

Algorithm

	For each measurement in the file	
		If this is the first measurement
			If it is from Radar
				Convert from polar coordinates to cartesian coordinates
			End If
			Initialize measurements
		Else
			If current timestamp is different from previous timestamp
				Predict the current state
			End If
			Update based on sensor type
		End If
	End For

Testing

• Using combined sensors (use_laser_ = true and use_radar_ = true):

  • RMSE (px, py, vx, and vy) = 0.078 - 0.085 - 0.215 - 0.268

• Using only laser (use_laser_ = true and use_radar_ = false):

  • RMSE (px, py, vx, and vy) = 0.170 - 0.145 - 0.269 - 0.333

• Using only radar (use_laser_ = false and use_radar_ = true):

  • RMSE (px, py, vx, and vy) = 0.228 - 0.341 - 0.550 - 1.002

• Using extended Kalman filter:

  • RMSE (px, py, vx, and vy) = 0.140 - 0.666 - 0.604 - 1.624

• NIS values were within the expected range (0.35 and 7.81)

• Log files are stored in the folder (output)