String Stable Constant-Spacing Platoon with Robustness to Communication Delays and Cutoff

Abstract

This dissertation develops a string stable, constant-spacing control strategy for connectedautonomous vehicle networks with robustness to communication delays and cutoff. It is well-known that decentralized constant-spacing strategies cannot achieve string stability, which requires the error to decrease as it propagates downstream. Introducing a centralized approach could achieve string stability in constant-spacing platoons. However, the benefits of introducing centralized control are questionable under communication limitations, like delays and cutoffs, which are common in modern traffic networks. Therefore, this thesis develops a new constant-spacing control approach that is robust to communication delays and cutoff. Moreover, this thesis develops conditions for guaranteeing string stability for constant-spacing policy (CSP) under communication delays. Furthermore, this dissertation applies the proposed control approach to a mixed vehicle network at a signalized intersection with vehicle-to-infrastructure (V2I) connectivity, as well as a human-led platoon.