Predictive cruise control of connected and autonomous vehicles via reinforcement learning

Weinan Gao, Adedapo Odekunle, Yunfeng Chen, Zhong Ping Jiang

Research output: Contribution to journalArticle

Abstract

Predictive cruise control concerns designing controllers for autonomous vehicles using the broadcasted information from the traffic lights such that the idle time around the intersection can be reduced. This study proposes a novel adaptive optimal control approach based on reinforcement learning to solve the predictive cruise control problem of a platoon of connected and autonomous vehicles. First, the reference velocity is determined for each autonomous vehicle in the platoon. Second, a data-driven adaptive optimal control algorithm is developed to estimate the gains of the desired distributed optimal controllers without the exact knowledge of system dynamics. The obtained controller is able to regulate the headway, velocity, and acceleration of each vehicle in a suboptimal sense. The goal of trip time reduction is achieved without compromising vehicle safety and passenger comfort. Numerical simulations are presented to validate the efficacy of the proposed methodology.

Original languageEnglish (US)
Pages (from-to)2849-2855
Number of pages7
JournalIET Control Theory and Applications
Volume13
Issue number17
DOIs
StatePublished - Nov 26 2019

Fingerprint

Cruise control
Autonomous Vehicles
Predictive Control
Reinforcement learning
Reinforcement Learning
Controller
Adaptive Control
Optimal Control
Controllers
Optimal Algorithm
Data-driven
System Dynamics
Control Algorithm
Efficacy
Control Problem
Safety
Intersection
Traffic
Numerical Simulation
Methodology

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Human-Computer Interaction
  • Computer Science Applications
  • Control and Optimization
  • Electrical and Electronic Engineering

Cite this

Predictive cruise control of connected and autonomous vehicles via reinforcement learning. / Gao, Weinan; Odekunle, Adedapo; Chen, Yunfeng; Jiang, Zhong Ping.

In: IET Control Theory and Applications, Vol. 13, No. 17, 26.11.2019, p. 2849-2855.

Research output: Contribution to journalArticle

Gao, Weinan ; Odekunle, Adedapo ; Chen, Yunfeng ; Jiang, Zhong Ping. / Predictive cruise control of connected and autonomous vehicles via reinforcement learning. In: IET Control Theory and Applications. 2019 ; Vol. 13, No. 17. pp. 2849-2855.
@article{6fa88dbcafde4e648faa8e9e7ee561a9,
title = "Predictive cruise control of connected and autonomous vehicles via reinforcement learning",
abstract = "Predictive cruise control concerns designing controllers for autonomous vehicles using the broadcasted information from the traffic lights such that the idle time around the intersection can be reduced. This study proposes a novel adaptive optimal control approach based on reinforcement learning to solve the predictive cruise control problem of a platoon of connected and autonomous vehicles. First, the reference velocity is determined for each autonomous vehicle in the platoon. Second, a data-driven adaptive optimal control algorithm is developed to estimate the gains of the desired distributed optimal controllers without the exact knowledge of system dynamics. The obtained controller is able to regulate the headway, velocity, and acceleration of each vehicle in a suboptimal sense. The goal of trip time reduction is achieved without compromising vehicle safety and passenger comfort. Numerical simulations are presented to validate the efficacy of the proposed methodology.",
author = "Weinan Gao and Adedapo Odekunle and Yunfeng Chen and Jiang, {Zhong Ping}",
year = "2019",
month = "11",
day = "26",
doi = "10.1049/iet-cta.2018.6031",
language = "English (US)",
volume = "13",
pages = "2849--2855",
journal = "IET Control Theory and Applications",
issn = "1751-8644",
publisher = "Institution of Engineering and Technology",
number = "17",

}

TY - JOUR

T1 - Predictive cruise control of connected and autonomous vehicles via reinforcement learning

AU - Gao, Weinan

AU - Odekunle, Adedapo

AU - Chen, Yunfeng

AU - Jiang, Zhong Ping

PY - 2019/11/26

Y1 - 2019/11/26

N2 - Predictive cruise control concerns designing controllers for autonomous vehicles using the broadcasted information from the traffic lights such that the idle time around the intersection can be reduced. This study proposes a novel adaptive optimal control approach based on reinforcement learning to solve the predictive cruise control problem of a platoon of connected and autonomous vehicles. First, the reference velocity is determined for each autonomous vehicle in the platoon. Second, a data-driven adaptive optimal control algorithm is developed to estimate the gains of the desired distributed optimal controllers without the exact knowledge of system dynamics. The obtained controller is able to regulate the headway, velocity, and acceleration of each vehicle in a suboptimal sense. The goal of trip time reduction is achieved without compromising vehicle safety and passenger comfort. Numerical simulations are presented to validate the efficacy of the proposed methodology.

AB - Predictive cruise control concerns designing controllers for autonomous vehicles using the broadcasted information from the traffic lights such that the idle time around the intersection can be reduced. This study proposes a novel adaptive optimal control approach based on reinforcement learning to solve the predictive cruise control problem of a platoon of connected and autonomous vehicles. First, the reference velocity is determined for each autonomous vehicle in the platoon. Second, a data-driven adaptive optimal control algorithm is developed to estimate the gains of the desired distributed optimal controllers without the exact knowledge of system dynamics. The obtained controller is able to regulate the headway, velocity, and acceleration of each vehicle in a suboptimal sense. The goal of trip time reduction is achieved without compromising vehicle safety and passenger comfort. Numerical simulations are presented to validate the efficacy of the proposed methodology.

UR - http://www.scopus.com/inward/record.url?scp=85075625766&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075625766&partnerID=8YFLogxK

U2 - 10.1049/iet-cta.2018.6031

DO - 10.1049/iet-cta.2018.6031

M3 - Article

AN - SCOPUS:85075625766

VL - 13

SP - 2849

EP - 2855

JO - IET Control Theory and Applications

JF - IET Control Theory and Applications

SN - 1751-8644

IS - 17

ER -