Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems

This work presents the design and implementation of a data-driven Nonlinear Model Predictive Control (NMPC) framework for an Uncrewed Aerial Vehicle (UAV) equipped with a 3-DOF robotic arm. Real-world data was collected using the Matrice 100 platform and Dynamixel MX-28AR actuators to identify a hig...

Full description

Saved in:
Bibliographic Details
Main Authors: Bryan S. Guevara, Jose Varela-Aldas, Viviana Moya, Manuel Cardona, Daniel C. Gandolfo, Juan M. Toibero
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11020657/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849727771609137152
author Bryan S. Guevara
Jose Varela-Aldas
Viviana Moya
Manuel Cardona
Daniel C. Gandolfo
Juan M. Toibero
author_facet Bryan S. Guevara
Jose Varela-Aldas
Viviana Moya
Manuel Cardona
Daniel C. Gandolfo
Juan M. Toibero
author_sort Bryan S. Guevara
collection DOAJ
description This work presents the design and implementation of a data-driven Nonlinear Model Predictive Control (NMPC) framework for an Uncrewed Aerial Vehicle (UAV) equipped with a 3-DOF robotic arm. Real-world data was collected using the Matrice 100 platform and Dynamixel MX-28AR actuators to identify a high-dimensional linear model via Dynamic Mode Decomposition with Control (DMDc), capturing the interactions between the aerial vehicle and the manipulator across 21 state variables. This DMDc-based model is embedded within the NMPC formulation to predict system behavior over finite horizons. The UAV’s orientation is represented using quaternions, enabling continuous and singularity-free attitude control. Additionally, the redundancy of the UAV-manipulator system allows for the integration of secondary objectives into the cost function, supporting flexible task execution. To meet real-time requirements, the control problem is solved using the Acados solver. The resulting controller achieves high-precision tracking while managing internal constraints, demonstrating the potential of data-driven NMPC in aerial manipulation tasks.
format Article
id doaj-art-b631940b72b942cf830f68c4932f1d6f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-b631940b72b942cf830f68c4932f1d6f2025-08-20T03:09:45ZengIEEEIEEE Access2169-35362025-01-0113968349684310.1109/ACCESS.2025.357572411020657Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator SystemsBryan S. Guevara0https://orcid.org/0000-0002-7830-0498Jose Varela-Aldas1https://orcid.org/0000-0002-4084-1424Viviana Moya2https://orcid.org/0000-0002-6064-6925Manuel Cardona3https://orcid.org/0000-0002-4211-3498Daniel C. Gandolfo4https://orcid.org/0000-0002-4938-2105Juan M. Toibero5https://orcid.org/0000-0001-9347-566XInstituto de Automática, Universidad Nacional de San Juan-CONICET, San Juan, ArgentinaCentro de Investigación MIST, Facultad de Ingenierías, Universidad Tecnológica Indoamérica, Ambato, EcuadorFacultad de Ciencias Técnicas, Universidad Internacional del Ecuador (UIDE), Quito, EcuadorResearch Department, Universidad Don Bosco, Soyapango, El SalvadorInstituto de Automática, Universidad Nacional de San Juan-CONICET, San Juan, ArgentinaInstituto de Automática, Universidad Nacional de San Juan-CONICET, San Juan, ArgentinaThis work presents the design and implementation of a data-driven Nonlinear Model Predictive Control (NMPC) framework for an Uncrewed Aerial Vehicle (UAV) equipped with a 3-DOF robotic arm. Real-world data was collected using the Matrice 100 platform and Dynamixel MX-28AR actuators to identify a high-dimensional linear model via Dynamic Mode Decomposition with Control (DMDc), capturing the interactions between the aerial vehicle and the manipulator across 21 state variables. This DMDc-based model is embedded within the NMPC formulation to predict system behavior over finite horizons. The UAV’s orientation is represented using quaternions, enabling continuous and singularity-free attitude control. Additionally, the redundancy of the UAV-manipulator system allows for the integration of secondary objectives into the cost function, supporting flexible task execution. To meet real-time requirements, the control problem is solved using the Acados solver. The resulting controller achieves high-precision tracking while managing internal constraints, demonstrating the potential of data-driven NMPC in aerial manipulation tasks.https://ieeexplore.ieee.org/document/11020657/Aerial manipulatorNMPCCasADiAcadosreal-time optimizationUAV dynamics
spellingShingle Bryan S. Guevara
Jose Varela-Aldas
Viviana Moya
Manuel Cardona
Daniel C. Gandolfo
Juan M. Toibero
Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
IEEE Access
Aerial manipulator
NMPC
CasADi
Acados
real-time optimization
UAV dynamics
title Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
title_full Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
title_fullStr Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
title_full_unstemmed Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
title_short Data-Driven Model Predictive Control for Trajectory Tracking in UAV-Manipulator Systems
title_sort data driven model predictive control for trajectory tracking in uav manipulator systems
topic Aerial manipulator
NMPC
CasADi
Acados
real-time optimization
UAV dynamics
url https://ieeexplore.ieee.org/document/11020657/
work_keys_str_mv AT bryansguevara datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems
AT josevarelaaldas datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems
AT vivianamoya datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems
AT manuelcardona datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems
AT danielcgandolfo datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems
AT juanmtoibero datadrivenmodelpredictivecontrolfortrajectorytrackinginuavmanipulatorsystems