Collaborators

• Giacomo Corradini
• Mattia Pettene

Autonomous driving simulator

The project consists in a semplified version of an autonomous driving system, in particular are developed the longitudinal control for a traffic-light intersection and a lateral control for obstacle avoidance. The path plannig is performed using the high-level planner RRT (rapidly-exploring random tree) The longitudinal control is performed using the motion primitive technique The lateral control is performed using the preview point lateral control (is also defined the clothoid based lateral control)

Clone repositorry

git clone --recurse-submodules  https://github.com/GiacomoCorradini/Autonomous_driving_simulator.git ~/Autonomous_driving_simulator

Compile agent communication library

1. enter inside the folder agent_communication_lib

cd ~/Autonomous_driving_simulator/agent_communication_lib

2. compile the project with cmake (need ruby)

cmake -Bbuild
cmake --build build -t install

3. copy file in bin/lib/libagent_lib.so into ../PyDrivingSim/agent
4. copy file in bin/lib_static/libagent_lib.a into ../basic_agent_st/lib

Compile Clothoids library

1. enter inside the folder Clothoids

cd ~/Autonomous_driving_simulator/basic_agent/Clothoids

2. compile the project with cmake

cmake -Bbuild
cmake --build build -t install

Compile basic_agent

1. enter inside the folder basic_agent

cd ~/Autonomous_driving_simulator/basic_agent

2. compile the project with cmake

cmake -Bbuild
cmake --build build

Run the project

Open two different terminals

Run basic_agent in the first terminal

1. enter inside the folder basic_agent

cd ~/Autonomous_driving_simulator/basic_agent

2. run basic_agent

./bin/basic_agent

Run pydriving simulator in the second terminal

1. enter inside the folder PyDrivingSim

cd ~/Autonomous_driving_simulator/PyDrivingSim

2. run simulator.py

• to simulate the traffic-light environment

python3 simulator.py 0

• to simulate the traffic-light environment + obstacle avoidance

python3 simulator.py 1