A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV
This paper investigates the development of a deep learning-based flight control model for a tilt-rotor unmanned aerial vehicle, focusing on altitude, speed, and roll hold systems. Training data is gathered from the X-Plane flight simulator, employing a proportional–integral–derivative controller to...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-01-01
|
Series: | Aerospace |
Subjects: | |
Online Access: | https://www.mdpi.com/2226-4310/11/1/96 |
_version_ | 1797344803541745664 |
---|---|
author | Javensius Sembiring Rianto Adhy Sasongko Eduardo I. Bastian Bayu Aji Raditya Rayhan Ekananto Limansubroto |
author_facet | Javensius Sembiring Rianto Adhy Sasongko Eduardo I. Bastian Bayu Aji Raditya Rayhan Ekananto Limansubroto |
author_sort | Javensius Sembiring |
collection | DOAJ |
description | This paper investigates the development of a deep learning-based flight control model for a tilt-rotor unmanned aerial vehicle, focusing on altitude, speed, and roll hold systems. Training data is gathered from the X-Plane flight simulator, employing a proportional–integral–derivative controller to enhance flight dynamics and data quality. The model architecture, implemented within the TensorFlow framework, undergoes iterative tuning for optimal performance. Testing involved two scenarios: wind-free conditions and wind disturbances. In wind-free conditions, the model demonstrated excellent tracking performance, closely tracking the desired altitude. The model’s robustness is further evaluated by introducing wind disturbances. Interestingly, these disturbances do not significantly impact the model performance. This research has demonstrated data-driven flight control in a tilt-rotor unmanned aerial vehicle, offering improved adaptability and robustness compared to traditional methods. Future work may explore further flight modes, environmental complexities, and the utilization of real test flight data to enhance the model generalizability. |
first_indexed | 2024-03-08T11:09:09Z |
format | Article |
id | doaj.art-2dc9ea46207d476887e12511ef15ba64 |
institution | Directory Open Access Journal |
issn | 2226-4310 |
language | English |
last_indexed | 2024-03-08T11:09:09Z |
publishDate | 2024-01-01 |
publisher | MDPI AG |
record_format | Article |
series | Aerospace |
spelling | doaj.art-2dc9ea46207d476887e12511ef15ba642024-01-26T14:14:21ZengMDPI AGAerospace2226-43102024-01-011119610.3390/aerospace11010096A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAVJavensius Sembiring0Rianto Adhy Sasongko1Eduardo I. Bastian2Bayu Aji Raditya3Rayhan Ekananto Limansubroto4Faculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, IndonesiaFaculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, IndonesiaFaculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, IndonesiaFaculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, IndonesiaFaculty of Mechanical and Aerospace Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, IndonesiaThis paper investigates the development of a deep learning-based flight control model for a tilt-rotor unmanned aerial vehicle, focusing on altitude, speed, and roll hold systems. Training data is gathered from the X-Plane flight simulator, employing a proportional–integral–derivative controller to enhance flight dynamics and data quality. The model architecture, implemented within the TensorFlow framework, undergoes iterative tuning for optimal performance. Testing involved two scenarios: wind-free conditions and wind disturbances. In wind-free conditions, the model demonstrated excellent tracking performance, closely tracking the desired altitude. The model’s robustness is further evaluated by introducing wind disturbances. Interestingly, these disturbances do not significantly impact the model performance. This research has demonstrated data-driven flight control in a tilt-rotor unmanned aerial vehicle, offering improved adaptability and robustness compared to traditional methods. Future work may explore further flight modes, environmental complexities, and the utilization of real test flight data to enhance the model generalizability.https://www.mdpi.com/2226-4310/11/1/96tilt-rotorflight controldeep learningdata-driven control model |
spellingShingle | Javensius Sembiring Rianto Adhy Sasongko Eduardo I. Bastian Bayu Aji Raditya Rayhan Ekananto Limansubroto A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV Aerospace tilt-rotor flight control deep learning data-driven control model |
title | A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV |
title_full | A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV |
title_fullStr | A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV |
title_full_unstemmed | A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV |
title_short | A Deep Learning Approach for Trajectory Control of Tilt-Rotor UAV |
title_sort | deep learning approach for trajectory control of tilt rotor uav |
topic | tilt-rotor flight control deep learning data-driven control model |
url | https://www.mdpi.com/2226-4310/11/1/96 |
work_keys_str_mv | AT javensiussembiring adeeplearningapproachfortrajectorycontroloftiltrotoruav AT riantoadhysasongko adeeplearningapproachfortrajectorycontroloftiltrotoruav AT eduardoibastian adeeplearningapproachfortrajectorycontroloftiltrotoruav AT bayuajiraditya adeeplearningapproachfortrajectorycontroloftiltrotoruav AT rayhanekanantolimansubroto adeeplearningapproachfortrajectorycontroloftiltrotoruav AT javensiussembiring deeplearningapproachfortrajectorycontroloftiltrotoruav AT riantoadhysasongko deeplearningapproachfortrajectorycontroloftiltrotoruav AT eduardoibastian deeplearningapproachfortrajectorycontroloftiltrotoruav AT bayuajiraditya deeplearningapproachfortrajectorycontroloftiltrotoruav AT rayhanekanantolimansubroto deeplearningapproachfortrajectorycontroloftiltrotoruav |