Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car

Mengzhe Huang; Mingyu Zhao; Parthiv Parikh; Yebin Wang; Kaan Ozbay; Zhong Ping Jiang

doi:10.1109/ICCA.2019.8900002

Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car

Mengzhe Huang, Mingyu Zhao, Parthiv Parikh, Yebin Wang, Kaan Ozbay, Zhong Ping Jiang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Lateral control design is one of the fundamental components for self-driving cars. In this paper, we propose a learning-based control strategy that enables a mobile car equipped with a camera to perfectly perform lane keeping in a road on the ground. Using the method of adaptive dynamic programming, the proposed control algorithm exploits the structural knowledge of the car kinematics as well as the collected data (images) about the lane information. An adaptive optimal lateral controller is obtained through a data-driven learning algorithm. The effectiveness of the proposed method is demonstrated by theoretical stability proofs and experimental evaluations.

Original language	English (US)
Title of host publication	2019 IEEE 15th International Conference on Control and Automation, ICCA 2019
Publisher	IEEE Computer Society
Pages	1126-1131
Number of pages	6
ISBN (Electronic)	9781728111643
DOIs	https://doi.org/10.1109/ICCA.2019.8900002
State	Published - Jul 2019
Event	15th IEEE International Conference on Control and Automation, ICCA 2019 - Edinburgh, United Kingdom Duration: Jul 16 2019 → Jul 19 2019

Publication series

Name	IEEE International Conference on Control and Automation, ICCA
Volume	2019-July
ISSN (Print)	1948-3449
ISSN (Electronic)	1948-3457

Conference

Conference	15th IEEE International Conference on Control and Automation, ICCA 2019
Country/Territory	United Kingdom
City	Edinburgh
Period	7/16/19 → 7/19/19

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Control and Systems Engineering
Electrical and Electronic Engineering
Industrial and Manufacturing Engineering

Access to Document

10.1109/ICCA.2019.8900002

Cite this

Huang, M., Zhao, M., Parikh, P., Wang, Y., Ozbay, K., & Jiang, Z. P. (2019). Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car. In 2019 IEEE 15th International Conference on Control and Automation, ICCA 2019 (pp. 1126-1131). Article 8900002 (IEEE International Conference on Control and Automation, ICCA; Vol. 2019-July). IEEE Computer Society. https://doi.org/10.1109/ICCA.2019.8900002

Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car. / Huang, Mengzhe; Zhao, Mingyu; Parikh, Parthiv et al.
2019 IEEE 15th International Conference on Control and Automation, ICCA 2019. IEEE Computer Society, 2019. p. 1126-1131 8900002 (IEEE International Conference on Control and Automation, ICCA; Vol. 2019-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Huang, M, Zhao, M, Parikh, P, Wang, Y, Ozbay, K & Jiang, ZP 2019, Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car. in 2019 IEEE 15th International Conference on Control and Automation, ICCA 2019., 8900002, IEEE International Conference on Control and Automation, ICCA, vol. 2019-July, IEEE Computer Society, pp. 1126-1131, 15th IEEE International Conference on Control and Automation, ICCA 2019, Edinburgh, United Kingdom, 7/16/19. https://doi.org/10.1109/ICCA.2019.8900002

@inproceedings{9bab1e0aefa6487ca01a2c349a0c011f,

title = "Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car",

abstract = "Lateral control design is one of the fundamental components for self-driving cars. In this paper, we propose a learning-based control strategy that enables a mobile car equipped with a camera to perfectly perform lane keeping in a road on the ground. Using the method of adaptive dynamic programming, the proposed control algorithm exploits the structural knowledge of the car kinematics as well as the collected data (images) about the lane information. An adaptive optimal lateral controller is obtained through a data-driven learning algorithm. The effectiveness of the proposed method is demonstrated by theoretical stability proofs and experimental evaluations.",

author = "Mengzhe Huang and Mingyu Zhao and Parthiv Parikh and Yebin Wang and Kaan Ozbay and Jiang, {Zhong Ping}",

note = "Funding Information: This work has been supported in part by the National Science Foundation under Grant ECCS-1501044, in part by a gift from the Mitsubishi Electric Research Laboratories and in part by C2SMART Center.; 15th IEEE International Conference on Control and Automation, ICCA 2019 ; Conference date: 16-07-2019 Through 19-07-2019",

year = "2019",

month = jul,

doi = "10.1109/ICCA.2019.8900002",

language = "English (US)",

series = "IEEE International Conference on Control and Automation, ICCA",

publisher = "IEEE Computer Society",

pages = "1126--1131",

booktitle = "2019 IEEE 15th International Conference on Control and Automation, ICCA 2019",

}

TY - GEN

T1 - Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car

AU - Huang, Mengzhe

AU - Zhao, Mingyu

AU - Parikh, Parthiv

AU - Wang, Yebin

AU - Ozbay, Kaan

AU - Jiang, Zhong Ping

N1 - Funding Information: This work has been supported in part by the National Science Foundation under Grant ECCS-1501044, in part by a gift from the Mitsubishi Electric Research Laboratories and in part by C2SMART Center.

PY - 2019/7

Y1 - 2019/7

N2 - Lateral control design is one of the fundamental components for self-driving cars. In this paper, we propose a learning-based control strategy that enables a mobile car equipped with a camera to perfectly perform lane keeping in a road on the ground. Using the method of adaptive dynamic programming, the proposed control algorithm exploits the structural knowledge of the car kinematics as well as the collected data (images) about the lane information. An adaptive optimal lateral controller is obtained through a data-driven learning algorithm. The effectiveness of the proposed method is demonstrated by theoretical stability proofs and experimental evaluations.

AB - Lateral control design is one of the fundamental components for self-driving cars. In this paper, we propose a learning-based control strategy that enables a mobile car equipped with a camera to perfectly perform lane keeping in a road on the ground. Using the method of adaptive dynamic programming, the proposed control algorithm exploits the structural knowledge of the car kinematics as well as the collected data (images) about the lane information. An adaptive optimal lateral controller is obtained through a data-driven learning algorithm. The effectiveness of the proposed method is demonstrated by theoretical stability proofs and experimental evaluations.

UR - http://www.scopus.com/inward/record.url?scp=85075799299&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075799299&partnerID=8YFLogxK

U2 - 10.1109/ICCA.2019.8900002

DO - 10.1109/ICCA.2019.8900002

M3 - Conference contribution

AN - SCOPUS:85075799299

T3 - IEEE International Conference on Control and Automation, ICCA

SP - 1126

EP - 1131

BT - 2019 IEEE 15th International Conference on Control and Automation, ICCA 2019

PB - IEEE Computer Society

T2 - 15th IEEE International Conference on Control and Automation, ICCA 2019

Y2 - 16 July 2019 through 19 July 2019

ER -

Reinforcement Learning for Vision-Based Lateral Control of a Self-Driving Car

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this