tsbertalan/carnd-capstone

System integration for highway driving with traffic lights

This is the project repo for the final project of the Udacity Self-Driving Car Nanodegree: Programming a Real Self-Driving Car. For more information about the project, see the project introduction here.

Please use one of the two installation options, either native or docker installation.

This project was completed by Dean Liu, Louis Chow, Luuthien Xuan, Prasad Pillai, and Tom Bertalan.

Native Installation

• Be sure that your workstation is running Ubuntu 16.04 Xenial Xerus or Ubuntu 14.04 Trusty Tahir. Ubuntu downloads can be found here.

• If using a Virtual Machine to install Ubuntu, use the following configuration as minimum:

  • 2 CPU
  • 2 GB system memory
  • 25 GB of free hard drive space

  The Udacity provided virtual machine has ROS and Dataspeed DBW already installed, so you can skip the next two steps if you are using this.

• Follow these instructions to install ROS

  • ROS Kinetic if you have Ubuntu 16.04.
  • ROS Indigo if you have Ubuntu 14.04.

• Dataspeed DBW

  • Use this option to install the SDK on a workstation that already has ROS installed: One Line SDK Install (binary)

• Download the Udacity Simulator.

• Either clone this repository to /capstone, or clone it elsewhere and symlink it there with ln -s "$CLONEDIR" /capstone.

• Make a python 2.7 virtual environment. virtualenvwrapper is great for this.

• If installing on a machine with a CUDA- and TensorFlow-capable GPU:

  • Follow TensorFlow's installation instructions, including the link to instructions for installing the CUDA toolkit and cuDNN. You may want to insert any commands for modifying your environment variables, such as LD_LIBRARY_PATH, into your ~/.profile file.
  • Change tensorflow to tensorflow-gpu in requirements.txt.
  • Note that CUDA 8 may be necessary for running with Python 2.7, as is the norm for ROS. This can be installed alongside CUDA 9 with sudo apt-get install cuda-8-0.

• Reproduce the installation steps from Dockerfile, including running tod_coco_install.sh. Note that you'll first need to run e.g. export ROS_DISTRO=kinetic depending on your ROS distribution. Note that tod_coco_install.sh modifies your .bashrc file.

• Still in the virtualenv, run pip uninstall em and pip install catkin_pkg em rospkg.

• In /capstone/ros, run catkin_make.

• Either in a terminal where roslaunch is to be run, or permanently via ~/.bashrc or ~/.profile, run source /capstone/ros/devel/setup.sh.

• Download luu_real.pb and luu_sim.pb and put them both in /capstone/ros/src/tl_detector/light_classification/saved_nets, making this directory if necessary.

• Optionally, install the the Rviz plugins with sudo apt-get install ros-kinetic-jsk-rviz-plugins libbullet-dev libsdl-image1.2-dev libsdl-dev.

Docker Installation

Install Docker

Build the docker container

docker build . -t capstone

Run the docker file

docker run -p 4567:4567 -v $PWD:/capstone -v /tmp/log:/root/.ros/ --rm -it capstone

Port Forwarding

To set up port forwarding, please refer to the instructions from term 2

Usage

In /capstone/ros, after establishing the right $PYTHONPATH environment variable to (likely via ~/.profile) and sourcing /capstone/ros/devel/setup.sh, run the controller and light-detector with roslaunch launch/styx.launch.

When the lauch has completed and all nodes are running, use run the simulator, and uncheck "Manual", and check "Camera".

Real world testing

1. Download training bag that was recorded on the Udacity self-driving car.
2. Unzip the file

unzip traffic_light_bag_file.zip

3. Play the bag file

rosbag play -l traffic_light_bag_file/traffic_light_training.bag

4. Launch your project in site mode

cd CarND-Capstone/ros
roslaunch launch/site.launch

5. Confirm that traffic light detection works on real life images