IMPLEMENTASI REINFORCEMENT LEARNING UNTUK STABILISASI SUDUT PITCH 90° PADA MODEL ROKET 6DOF DI MATLAB SIMULINK

*Cornelius Gian  -  Department of Mechanical Engineering, Universitas Diponegoro, Jl. Prof. Sudarto, SH, Tembalang, Semarang, Indonesia 50275, Indonesia
Mochammad Ariyanto  -  Department of Mechanical Engineering, Universitas Diponegoro, Jl. Prof. Sudarto, SH, Tembalang, Semarang, Indonesia 50275, Indonesia
Joga Dharma Setiawan  -  Department of Mechanical Engineering, Universitas Diponegoro, Jl. Prof. Sudarto, SH, Tembalang, Semarang, Indonesia 50275, Indonesia

Citation Format:
Abstract

Perkembangan teknologi roket modern menuntut sistem kontrol yang mampu menjaga kestabilan roket secara presisi. Tantangan muncul akibat dinamika roket yang bersifat nonlinear dan kompleks sehingga metode kontrol konvensional kurang efektif. Penelitian ini bertujuan mengimplementasikan algoritma Reinforcement Learning (RL) khususnya Twin Delayed Deep Deterministic Policy Gradient (TD3), untuk mengendalikan defleksi fin dalam menstabilkan sudut pitch 90° pada model roket 6DoF di MATLAB Simulink. Metode penelitian meliputi persiapan model Simulink roket 6DoF, desain fungsi reward, pembuatan environment RL, pelatihan agen RL, serta pengujian performa agen melalui simulasi dengan gangguan angin. Hasil penelitian menunjukkan bahwa pada sistem tanpa RL, nilai Mean Absolute Error (MAE) untuk gain 1, 2, 3, 4, dan 5 berturut-turut adalah sebesar 0.6242°, 1.2483°, 1.8719°, 2.4949°, dan 3.1172°. Setelah implementasi RL, nilai MAE menurun menjadi 0.2770°, 0.3738°, 0.4351°, 1.2211°, dan 2.1156°. Sistem dengan RL menunjukkan peningkatan akurasi kontrol pitch.  Hal ini membuktikan bahwa agen RL TD3 mampu mengatasi dinamika roket yang kompleks secara adaptif.

Fulltext View|Download
Keywords: defleksi fin, kontrol pitch, reinforcement learning, roket 6dof, td3
Article Info
Section: Articles
Language : ID
Recent articles
ANALISIS THERMOELASTO-HYDRODYNAMIC PADA MULTILAYER GAS FOIL THRUST BEARING DENGAN VARIASI MATERIAL PELAPIS MENGGUNAKAN PENDEKATAN FLUID STRUCTURE INTERACTION(FSI) ANALISIS PENGARUH DIAMETER NOZZLE SAAT EKSTRUSI PADA HASIL SERAT POLYCAPROLACTONE MENGGUNAKAN METODE WET SPINNING UJI DENSITAS PELET PENYERAP RAMAH LINGKUNGAN DARI AMPAS TAHU KOMBINASI DENGAN BUBUK DAUN PANDAN More recent articles
  1. Ferro C, Cafaro M, Maggiore P. Optimizing Solid Rocket Missile Trajectories: A Hybrid Approach Using an Evolutionary Algorithm and Machine Learning. Aerospace 2024;11. https://doi.org/10.3390/aerospace11110912
  2. Ferrante Reuben. A Robust Control Approach for Rocket Landing 2017:1–78
  3. Jiang Y, Yang Y, Lan Z, Zhan G, Li SE, Sun Q, et al. Rocket Landing Control with Random Annealing Jump Start Reinforcement Learning 2024
  4. Brötje S, Kirchner M, Giovannetti F. Performance and heat transfer analysis of uncovered photovoltaic-thermal collectors with detachable compound. Sol Energy 2018;170:406–18. https://doi.org/10.1016/j.solener.2018.05.030
  5. Wada D, Araujo-Estrada SA, Windsor S. Unmanned Aerial Vehicle Pitch Control Using Deep Reinforcement Learning with Discrete Actions in Wind Tunnel Test. Aerospace 2021. https://doi.org/https://doi.org/10.3390/aerospace8010018
  6. Kisabo AB, Adebimpe AF, Samuel SO. Pitch Control of a Rocket with a Novel LQG/LTR Control Algorithm. J Aircr Spacecr Technol 2019;3:24–37. https://doi.org/10.3844/jastsp.2019.24.37
  7. Xue S, Wang Z, Bai H, Yu C, Li Z. Research on Self-Learning Control Method of Reusable Launch Vehicle Based on Neural Network Architecture Search. Aerospace 2024;11. https://doi.org/10.3390/aerospace11090774
  8. Putro IE, Subiantoro A, Halim A, Triharjanto RH, Syafiie S. Optimal Control Design of Slow Dominant Transient Response for Longitudinal Missile Dynamics. 2023 IEEE Int Conf Aerosp Electron Remote Sens Technol ICARES 2023 2023:1–8. https://doi.org/10.1109/ICARES60489.2023.10329801
  9. Iafrate D, Brandonisio A, Hinz R, Lavagna M. Propulsive landing of launchers’ first stages with Deep Reinforcement Learning. Acta Astronaut 2025;227:40–56. https://doi.org/10.1016/j.actaastro.2024.11.028
  10. Kisabo AB, Adebimpe AF, Okwo OC, Samuel SO. State-Space Modelling of a Rocket for Optimal Control System Design. J Aircr Spacecr Technol 2019;3:128–37. https://doi.org/10.3844/jastsp.2019.128.137
  11. Kim S-H, Lee Y-I, Tahk M-J. New Structure for an Aerodynamic Fin Control System for Tail Fin-Controlled STT Missiles. J Aerosp Eng 2011;24:505–10. https://doi.org/10.1061/(asce)as.1943-5525.0000088
  12. Chen Y, Ma L. Rocket powered landing guidance using proximal policy optimization. ACM Int Conf Proceeding Ser 2019. https://doi.org/10.1145/3351917.3351935
  13. Srinivasan A. Reinforcement Learning: Advancements, Limitations, and Real-world Applications. Interantional J Sci Res Eng Manag 2023;07. https://doi.org/10.55041/ijsrem25118
  14. Tevera-Ruiz A, Garcia-Rodriguez R, Parra-Vega V, Ramos-Velasco LE. Q-Learning with the Variable Box Method: A Case Study to Land a Solid Rocket. Machines 2023;11:1–14. https://doi.org/10.3390/machines11020214

Last update:

No citation recorded.

Last update:

No citation recorded.