SelfDrivingCarsControlDesign Save

Self Driving Cars Longitudinal and Lateral Control Design

Project README

Self Driving Car Control Design

Self Driving Cars Longitudinal and Lateral Control Design

In this project, I implemented a controller in Python and use it to drive a car autonomously around a track in Carla Simulator.

Project notes:

The waypoints include positions as well as the speed the vehicle should attain. As a result, the waypoints become the reference signal for our controller and navigating to all the waypoints effectively completes the full track.

Since the controller reference contains both position and speed, we need to implement both Longitudinal and Lateral Control.

The output of the controller will be the vehicle throttle, brake and steering angle commands.

The throttle and brake come from the Longitudinal speed control and the steering comes from our Lateral Control.

If you want to make your own controller design, I highly recommend that you should consider the following points:

If you open the simulator directory and navigate to the Course1FinalProject folder, you will see a file named controller2d.py

These controller functions are as the following:

Screenshot_6

You can adjust and re-create these functions according to your own custom controller design.

The Race Tack including waypoints:

Screenshot_4

Some concepts including formulas and expected output of the controller :

Screenshot_1

Screenshot_2

Screenshot_3

Screenshot from the application:

Screenshot_7

Final Notes:

This poject is my own implementation of the project assignment of Self Driving Cars Specialization offered by University of Toronto on Coursera.

  • I added Carla Simulator Installation guides for Ubuntu and Windows into the Guides folder.

  • Both in Windows and Ubuntu, add Course1FinalProject folder into the CarlaSimulator/PythonClient folder.

Use different terminals for the client and server.

How to run in Ubuntu:

  • Server terminal:

$ cd $HOME /opt/CarlaSimulator # where the CarlaSimulator is located

$ ./CarlaUE4.sh /Game/Maps/RaceTrack -windowed -carla-server -ResX=640 -ResY=480 -benchmark -fps=20

  • Client terminal:

$ cd $HOME /opt/CarlaSimulator/PythonClient/Course1FinalProject

$ python3 module_7.py

How to run in Windows:

  • Server terminal:

\> C:

\> cd \Coursera\CarlaSimulator

\> CarlaUE4.exe /Game/Maps/RaceTrack -windowed -carla-server -benchmark -fps=15 -ResX=640 -ResY=480

  • Client terminal:

\> C:

\> cd \Coursera\CarlaSimulator\PythonClient\Course1FinalProject

\> python module_7.py

Final Reminders for troubleshooting tips:

  • Python 3.7 is not currently compatible with CARLA. Use Python 3.5.x or Python 3.6.x in Windows or Ubuntu.

  • Allow CarlaUE4 to access through the Windows firewall if prompted to do so.

  • CARLA requires networking enabled with the firewall allowing access to the CARLA loader, and by default port 2000, 2001 and 2002 (TCP and UDP) available on the network. When you first run CARLA in server mode, Windows will prompt you to allow the application to access these ports if they are not already accessible on your system. If your network does not provide access to port 2000, please look at FAQ section in the installation guides in the Guides folder.

Open Source Agenda is not affiliated with "SelfDrivingCarsControlDesign" Project. README Source: enginBozkurt/SelfDrivingCarsControlDesign

Open Source Agenda Badge

Open Source Agenda Rating