Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case

Diva Kartika Larasati; Larasmoyo Nugroho; Sastra Kusuma Wijaya; Rika Andiarti; Rini Akmeliawati; Prawito Prajitno; Ery Fitrianingsih

doi:10.1109/ICARES56907.2022.9992304

Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case

Diva Kartika Larasati, Larasmoyo Nugroho, Sastra Kusuma Wijaya, Rika Andiarti, Rini Akmeliawati, Prawito Prajitno, Ery Fitrianingsih

Department of Physics

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

1 Citation (Scopus)

Abstract

A reward function in reinforcement learning is the formalization of the objective. Finding the ideal reward function is a challenge, that needs a search strategy to be constructed. Genetic Algorithm is a suitable approach for reward function search due to its thoroughness. The Deep Deterministic Policy Gradient (DDPG) algorithm, which is the focus of this research, is a reinforcement learning-based controller which performances are improved after the Genetic Algorithms optimizes the agent's reward functions. The optimized controller results in narrower missed distance and lower landing velocity compared to referenced DDPG controller, and significantly less fuel consumption compared to PID.

Original language	English
Title of host publication	2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665461917
DOIs	https://doi.org/10.1109/ICARES56907.2022.9992304
Publication status	Published - 2022
Event	2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Virtual, Online, Indonesia Duration: 24 Nov 2022 → 25 Nov 2022

Publication series

Name	2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings

Conference

Conference	2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022
Country/Territory	Indonesia
City	Virtual, Online
Period	24/11/22 → 25/11/22

Keywords

DDPG
reinforcement learning
reward shaping function
rocket landing

Access to Document

10.1109/ICARES56907.2022.9992304

Cite this

Larasati, D. K., Nugroho, L., Wijaya, S. K., Andiarti, R., Akmeliawati, R., Prajitno, P., & Fitrianingsih, E. (2022). Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case. In 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings (2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICARES56907.2022.9992304

Larasati, Diva Kartika ; Nugroho, Larasmoyo ; Wijaya, Sastra Kusuma et al. / Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case. 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings).

@inproceedings{824904d130d34c49a60814ab722d154c,

title = "Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case",

abstract = "A reward function in reinforcement learning is the formalization of the objective. Finding the ideal reward function is a challenge, that needs a search strategy to be constructed. Genetic Algorithm is a suitable approach for reward function search due to its thoroughness. The Deep Deterministic Policy Gradient (DDPG) algorithm, which is the focus of this research, is a reinforcement learning-based controller which performances are improved after the Genetic Algorithms optimizes the agent's reward functions. The optimized controller results in narrower missed distance and lower landing velocity compared to referenced DDPG controller, and significantly less fuel consumption compared to PID.",

keywords = "DDPG, reinforcement learning, reward shaping function, rocket landing",

author = "Larasati, {Diva Kartika} and Larasmoyo Nugroho and Wijaya, {Sastra Kusuma} and Rika Andiarti and Rini Akmeliawati and Prawito Prajitno and Ery Fitrianingsih",

note = "Funding Information: This research is partially supported by Rumah Program BRIN ORPA 2022 awarded to L. N. (B.20). Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 ; Conference date: 24-11-2022 Through 25-11-2022",

year = "2022",

doi = "10.1109/ICARES56907.2022.9992304",

language = "English",

series = "2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings",

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

booktitle = "2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings",

address = "United States",

}

Larasati, DK, Nugroho, L, Wijaya, SK, Andiarti, R, Akmeliawati, R, Prajitno, P & Fitrianingsih, E 2022, Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case. in 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings. 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings, Institute of Electrical and Electronics Engineers Inc., 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022, Virtual, Online, Indonesia, 24/11/22. https://doi.org/10.1109/ICARES56907.2022.9992304

Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case. / Larasati, Diva Kartika; Nugroho, Larasmoyo; Wijaya, Sastra Kusuma et al.
2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2022. (2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings).

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

TY - GEN

T1 - Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case

AU - Larasati, Diva Kartika

AU - Nugroho, Larasmoyo

AU - Wijaya, Sastra Kusuma

AU - Andiarti, Rika

AU - Akmeliawati, Rini

AU - Prajitno, Prawito

AU - Fitrianingsih, Ery

PY - 2022

Y1 - 2022

N2 - A reward function in reinforcement learning is the formalization of the objective. Finding the ideal reward function is a challenge, that needs a search strategy to be constructed. Genetic Algorithm is a suitable approach for reward function search due to its thoroughness. The Deep Deterministic Policy Gradient (DDPG) algorithm, which is the focus of this research, is a reinforcement learning-based controller which performances are improved after the Genetic Algorithms optimizes the agent's reward functions. The optimized controller results in narrower missed distance and lower landing velocity compared to referenced DDPG controller, and significantly less fuel consumption compared to PID.

AB - A reward function in reinforcement learning is the formalization of the objective. Finding the ideal reward function is a challenge, that needs a search strategy to be constructed. Genetic Algorithm is a suitable approach for reward function search due to its thoroughness. The Deep Deterministic Policy Gradient (DDPG) algorithm, which is the focus of this research, is a reinforcement learning-based controller which performances are improved after the Genetic Algorithms optimizes the agent's reward functions. The optimized controller results in narrower missed distance and lower landing velocity compared to referenced DDPG controller, and significantly less fuel consumption compared to PID.

KW - DDPG

KW - reinforcement learning

KW - reward shaping function

KW - rocket landing

UR - http://www.scopus.com/inward/record.url?scp=85146434044&partnerID=8YFLogxK

U2 - 10.1109/ICARES56907.2022.9992304

DO - 10.1109/ICARES56907.2022.9992304

M3 - Conference contribution

AN - SCOPUS:85146434044

T3 - 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings

BT - 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022

Y2 - 24 November 2022 through 25 November 2022

ER -

Larasati DK, Nugroho L, Wijaya SK, Andiarti R, Akmeliawati R, Prajitno P et al. Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case. In 2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2022. (2022 IEEE International Conference on Aerospace Electronics and Remote Sensing Technology, ICARES 2022 - Proceedings). doi: 10.1109/ICARES56907.2022.9992304

Genetic Algorithms Optimization of a Reinforcement Learning-based Controller for Vertical Landing Rocket Case

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this