ASDVC - A Self-Driving Vehicle Controller using Unsupervised Machine Learning

ASDVC is a self-driving vehicle controller that uses unsupervised machine learning methods, namely clustering based k-means, hierarchical, Gaussian Matrix Model and self-organizing mapping to optimize the path the vehicle follows from the source to destination. The real-time optimal selection of the unsupervised machine learning based motion control algorithm could provide fast response times of under one microsecond during the lateral, longitudinal and angular motion control of the autonomous vehicle. However, it is shown that a simple selection of one of the machine learning methods may not guarantee the optimality of the results. The successful implementation of ASDVC controller in self-driving vehicles could have a significant contribution towards making mobility more reliable and sustainable for the future vehicular transportation systems.