Design and Implementation of a Fully-Actuated Integrated Aerial Platform Based on Geometric Model Predictive Control

Area of Science:

• Robotics and Control Systems
• Aerospace Engineering
• Multi-Agent Systems

Background:

• Individual unmanned aerial vehicles (UAVs) often face limitations like underactuation and restricted maneuverability.
• Integrated Aerial Platforms (IAPs) composed of multiple UAVs offer enhanced payload, fault tolerance, and omnidirectional six-dimensional movement.
• These characteristics make multi-UAV IAPs suitable for advanced applications requiring integrated response, observation, and strike capabilities.

Purpose of the Study:

• To design and validate a novel integrated aerial platform (IAP) structure using a star-like configuration of three sub-UAVs.
• To develop and implement a robust control system for the IAP, enabling simultaneous six-dimensional position and attitude adjustments.
• To demonstrate the effectiveness of the proposed IAP design and control methodology through simulations and real-world prototype testing.

Main Methods:

• Modeling the dynamics of the integrated aerial platform (IAP) system.
• Designing a cascaded control system with an inner-loop subsystem for sub-UAV rotational motion and an outer-loop subsystem for overall IAP motion.
• Implementing a model predictive control on the manifold for the outer-loop subsystem to address non-Euclidean configuration spaces and constraints.
• Utilizing fieldbus technology for a real-time, scalable communication architecture among sub-UAVs.
• Developing a principle prototype of the multi-UAV IAP.

Main Results:

• The proposed IAP design enables omnidirectional six-dimensional movement and simultaneous adjustment of position and attitude.
• The developed control strategy ensures system stability and convergence to desired configurations, even with large initial errors.
• The IAP system successfully completed test scenarios that were unachievable by a baseline PID controller.
• The control scheme effectively handled state and input constraints, demonstrating its robustness.

Conclusions:

• The novel star-like configuration and model predictive control on the manifold provide a robust and effective solution for multi-UAV integrated aerial platforms.
• The developed IAP system demonstrates superior performance and maneuverability compared to traditional control methods, particularly in challenging scenarios.
• The successful validation through prototype testing confirms the practical applicability and potential of this integrated aerial platform technology.