A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving

Mei Lin Huang; Ching Hung Lee; Cheng Ting Huang; Hsin Han Chiang

doi:10.1109/APSIPAASC65261.2025.11249012

A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving

Mei Lin Huang
, Ching Hung Lee
, Cheng Ting Huang
, Hsin Han Chiang

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

Abstract

This paper presents a data-driven control framework for autonomous driving based on deep reinforcement learning (DRL). The proposed framework addresses lane-following and car-following using a dual-actor DDPG architecture with taskspecific rewards. The proposed approach improves safety, efficiency, and robustness, as demonstrated through extensive simulation, hardware-in-the-loop (HIL), and real-world testing. Comparative results show significant advantages over adaptive model predictive control (AMPC), highlighting its potential for real-world autonomous vehicle deployment.

Original language	English
Title of host publication	2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1338-1343
Number of pages	6
ISBN (Electronic)	9798331572068
DOIs	https://doi.org/10.1109/APSIPAASC65261.2025.11249012
Publication status	Published - 2025
Externally published	Yes
Event	17th Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025 - Singapore, Singapore Duration: 2025 Oct 22 → 2025 Oct 24

Publication series

Name	2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025

Conference

Conference	17th Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025
Country/Territory	Singapore
City	Singapore
Period	2025/10/22 → 2025/10/24

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Hardware and Architecture
Signal Processing

Access to Document

10.1109/APSIPAASC65261.2025.11249012

Cite this

Huang, M. L., Lee, C. H., Huang, C. T., & Chiang, H. H. (2025). A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving. In 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025 (pp. 1338-1343). (2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/APSIPAASC65261.2025.11249012

A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving. / Huang, Mei Lin; Lee, Ching Hung; Huang, Cheng Ting et al.
2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025. Institute of Electrical and Electronics Engineers Inc., 2025. p. 1338-1343 (2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025).

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

Huang, ML, Lee, CH, Huang, CT & Chiang, HH 2025, A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving. in 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025. 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025, Institute of Electrical and Electronics Engineers Inc., pp. 1338-1343, 17th Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025, Singapore, Singapore, 2025/10/22. https://doi.org/10.1109/APSIPAASC65261.2025.11249012

Huang ML, Lee CH, Huang CT, Chiang HH. A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving. In 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025. Institute of Electrical and Electronics Engineers Inc. 2025. p. 1338-1343. (2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025). doi: 10.1109/APSIPAASC65261.2025.11249012

Huang, Mei Lin ; Lee, Ching Hung ; Huang, Cheng Ting et al. / A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving. 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025. Institute of Electrical and Electronics Engineers Inc., 2025. pp. 1338-1343 (2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025).

@inproceedings{c6da5c3da0b44bbd9dbdd0c2b42627b2,

title = "A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving",

abstract = "This paper presents a data-driven control framework for autonomous driving based on deep reinforcement learning (DRL). The proposed framework addresses lane-following and car-following using a dual-actor DDPG architecture with taskspecific rewards. The proposed approach improves safety, efficiency, and robustness, as demonstrated through extensive simulation, hardware-in-the-loop (HIL), and real-world testing. Comparative results show significant advantages over adaptive model predictive control (AMPC), highlighting its potential for real-world autonomous vehicle deployment.",

author = "Huang, \{Mei Lin\} and Lee, \{Ching Hung\} and Huang, \{Cheng Ting\} and Chiang, \{Hsin Han\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 17th Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025 ; Conference date: 22-10-2025 Through 24-10-2025",

year = "2025",

doi = "10.1109/APSIPAASC65261.2025.11249012",

language = "English",

series = "2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1338--1343",

booktitle = "2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025",

}

TY - GEN

T1 - A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving

AU - Huang, Mei Lin

AU - Lee, Ching Hung

AU - Huang, Cheng Ting

AU - Chiang, Hsin Han

PY - 2025

Y1 - 2025

N2 - This paper presents a data-driven control framework for autonomous driving based on deep reinforcement learning (DRL). The proposed framework addresses lane-following and car-following using a dual-actor DDPG architecture with taskspecific rewards. The proposed approach improves safety, efficiency, and robustness, as demonstrated through extensive simulation, hardware-in-the-loop (HIL), and real-world testing. Comparative results show significant advantages over adaptive model predictive control (AMPC), highlighting its potential for real-world autonomous vehicle deployment.

AB - This paper presents a data-driven control framework for autonomous driving based on deep reinforcement learning (DRL). The proposed framework addresses lane-following and car-following using a dual-actor DDPG architecture with taskspecific rewards. The proposed approach improves safety, efficiency, and robustness, as demonstrated through extensive simulation, hardware-in-the-loop (HIL), and real-world testing. Comparative results show significant advantages over adaptive model predictive control (AMPC), highlighting its potential for real-world autonomous vehicle deployment.

UR - https://www.scopus.com/pages/publications/105030471455

UR - https://www.scopus.com/pages/publications/105030471455#tab=citedBy

U2 - 10.1109/APSIPAASC65261.2025.11249012

DO - 10.1109/APSIPAASC65261.2025.11249012

M3 - Conference contribution

AN - SCOPUS:105030471455

T3 - 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025

SP - 1338

EP - 1343

BT - 2025 Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 17th Asia Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2025

Y2 - 22 October 2025 through 24 October 2025

ER -

A Data-Driven Control Framework Using Deep Reinforcement Learning for Autonomous Driving

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this