This is the project repository for the Autonomous Vehicle Driving course @ Unisa.
Group 16
- Salvatore Ventre
- Vincenzo Russomanno
- Giovanni Puzo
- Vittorio Fina
Integrate a traffic light detector (already supplied, but not integrated) into a baseline and design a hierarchical planner that allows the vehicle to navigate autonomously within the Town01 map provided by the Carla simulator.
Behavioral planners and Local planners will need to be designed and implemented in a way that manage the behavior of the vehicle in the event of a red, yellow or green traffic light, as well as to avoid collisions with other vehicles and pedestrians present within the scene.
This software is made relying on the Python language and tested with the Carla Simulator in different scenarios. It is necessary to respect some dependencies, which for simplicity we have reported in a text file to be installed with the command:
/usr/bin/python3.6 -m pip install -r requirements.txt
First of all it is necessary to clone the repository in the PythonClient directory in Carla Simulator, by running these commands:
cd ../../CarlaSimulator/PythonClient/
git clone https://github.com/dev-guys-unisa/contest-avd-2021
Here are some fundamental indications to launch a demo of the project:
- Linux:
../../CarlaUE4.sh /Game/Maps/Town01 -carla-server -windowed -benchmark -fps=30 -quality-level=Epic
- Windows:
..\..\CarlaUE4.exe /Game/Maps/Town01 -carla-server -windowed -benchmark -fps=30 -quality-level=Epic
You have to move in src/
folder of the previous cloned repository and to launch main.py
file. Here there are the commands for Linux and Windows:
cd ../../CarlaSimulator/PythonClient/contest-avd-2021/src/
python3.6 main.py
IMPORTANT: If you want to launch only the client side, you have to specify the ip address and number of port of the Server in which Carla is running.
cd ../../CarlaSimulator/PythonClient/contest-avd-2021/src/
python3.6 main.py --host *ipaddress* --port *port*