Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A new platooning model for connected and autonomous vehicles to improve string stability

Shouwei Hui, Michael Zhang

TL;DR

The paper tackles reducing communication demands in CAV platoons while preserving or improving string stability. It proposes a leader-only platoon control (P-OVM) and a transition-phase variant (T-OVM) that adds local safety, with mathematical proofs showing P-OVM is always linearly stable and a stability criterion for T-OVM. Through ring-road and infinite-road simulations, the authors demonstrate that look-to-the-leader strategies outperform traditional multi-vehicle follower schemes in damping disturbances. This work offers a practically scalable approach to platooning that minimizes inter-vehicle communication without sacrificing stability, with potential impact on real-world CAV deployments.

Abstract

This paper presents a novel approach to coordinated vehicle platooning, where the platoon followers communicate solely with the platoon leader. A dynamic model is proposed to account for driving safety under communication delays. General linear stability results are mathematically proven, and numerical simulations are performed to analyze the impact of model parameters in two scenarios: a ring road with initial disturbance and an infinite road with periodic disturbance. The simulation outcomes align with the theoretical analysis, demonstrating that the proposed "look-to-the-leader" platooning strategy significantly outperforms conventional car-following strategies, such as following one or two vehicles ahead, in terms of traffic flow stabilization. This paper introduces a new perspective on organizing platoons for autonomous vehicles, with implications for enhancing traffic stability.

A new platooning model for connected and autonomous vehicles to improve string stability

TL;DR

The paper tackles reducing communication demands in CAV platoons while preserving or improving string stability. It proposes a leader-only platoon control (P-OVM) and a transition-phase variant (T-OVM) that adds local safety, with mathematical proofs showing P-OVM is always linearly stable and a stability criterion for T-OVM. Through ring-road and infinite-road simulations, the authors demonstrate that look-to-the-leader strategies outperform traditional multi-vehicle follower schemes in damping disturbances. This work offers a practically scalable approach to platooning that minimizes inter-vehicle communication without sacrificing stability, with potential impact on real-world CAV deployments.

Abstract

This paper presents a novel approach to coordinated vehicle platooning, where the platoon followers communicate solely with the platoon leader. A dynamic model is proposed to account for driving safety under communication delays. General linear stability results are mathematically proven, and numerical simulations are performed to analyze the impact of model parameters in two scenarios: a ring road with initial disturbance and an infinite road with periodic disturbance. The simulation outcomes align with the theoretical analysis, demonstrating that the proposed "look-to-the-leader" platooning strategy significantly outperforms conventional car-following strategies, such as following one or two vehicles ahead, in terms of traffic flow stabilization. This paper introduces a new perspective on organizing platoons for autonomous vehicles, with implications for enhancing traffic stability.
Paper Structure (12 sections, 3 theorems, 12 equations, 17 figures, 1 table)

This paper contains 12 sections, 3 theorems, 12 equations, 17 figures, 1 table.

Table of Contents

  1. Introduction
  2. The Non-Platoon and Platoon CF Models
  3. The optimal velocity model
  4. Platoon controller: A coordinated car-following strategy
  5. The transition phase model
  6. Stability analysis of the platoon models
  7. Numerical simulations
  8. Ring road with initial disturbance
  9. Infinite road with periodic disturbance
  10. Conclusions
  11. Proof of Theorem 3.1
  12. Proof of Theorem 3.3

Key Result

Theorem 3.1

The optimal velocity model OVM1 is linearly stable if

Figures (17)

  • Figure 1: Plot of an optimal velocity function with $l=5$m and $v_{\max}=20$m/s and the corresponding fundamental diagram.
  • Figure 2: Illustration of the platoon-controlled car-following system. Leading CAV is connected to all followers of the platoon and followers are directly following the leading CAV
  • Figure 3: Illustration of the transition phase car-following system. Leading CAV is connected to all followers inside the platoon and followers are following the leading CAV and the front CAV simultaneously
  • Figure 4: neutral stability lines of the transition phase model for selected percentages of leading vehicle sensitivity $b/(a+b)=0\%, 20\%, 40\%, 60\%, 80\%$. Above each line the region is stable and below is unstable.
  • Figure 5: Headway profile of OVM with sensitivity constant $a=0.4,\;0.8\;,1.6,\;2.4$
  • ...and 12 more figures

Theorems & Definitions (3)

  • Theorem 3.1
  • Theorem 3.2
  • Theorem 3.3