This repository contains code for a simulation that shows a bus traveling along a trajectory in an urban environment. The bus uses a specified control algorithm to follow the trajectory. Implemented control algorithms include Stanley control, Pure Pursuit control, and some variations.
I used this simulation in my first research project where I compared the effectiveness of multiple lateral control algorithms in driving an autonomous bus around a track in an urban environment. The article for this research is available at https://emerginginvestigators.org/articles/22-104. See Abstract.md.
Feel free to use this code for your own research or experimenting, or to see implementations of the Stanley and Pure Pursuit control algorithms.
results/: Contains excel spreadsheets of results.
tracks/: Contains waypoint files to be used.
course_images/: Contains diagrams of the courses used in the simulation.
simulation.py: Contains the simulation class used with functions for steering.
main.py: Contains the code which runs all control algorithms, or just one, and collects results in csv format.
lateral_controllers.py: Contains implementations of the control algorithms used.
all.sh: A script to run the simulation across all courses and all algorithms and all speeds.
Create virtual environment:
python -m venv .venv
source .venv/bin/activate
To run install requirements first:
pip install -r requirements.txt
Then run each file, e.g.
python main.py
