AROSV - An ROS based Self-Driving Vehicle Controller using Unsupervised Machine Learning Methods

B. Abegaz
Loyola University Chicago,
United States

Keywords: robot operating system (ROS), unsupervised machine learning, feedback control


AROSV is a robot operating system (ROS) based self-driving vehicle controller system that is designed to observe and control the movement of an autonomous vehicle from its starting position to the desired destination. Various computational and control mechanisms were implemented on the AROSV system using a closed-loop feedback motion controller and four unsupervised machine learning-based motion controllers. The proposed unsupervised machine learning-based motion control methods provide quicker response times of under one second during the lateral, longitudinal and angular motion control of the autonomous vehicle. The implementation of such methods could contribute to minimizing traffic congestion and avoiding collision for future vehicular transportation systems.