Intelligent control and path planning for unmanned aerial vehicles

Along with high-tech advances, unmanned aerial vehicles (UAVs) or drones are becoming popular in modern life. This technological trend will be expedited with UAVs being smartly employed in numerous human activities owing to the increasing advancement of learning systems. Their application also poses some venues in the control and planning of UAVs that have not been fully explored. This thesis is devoted to the development of intelligent control and path planning algorithms tailored for UAVs. Firstly, the modeling of multi-rotor aerial vehicles is refined by using the Newton-Euler method, incorporating vehicle dynamics to lay the foundation for displaying the responses of UAVs in interactions with the flight environment. This is important for the emerging technology of “digital twin” for UAV applications. Leveraging digital twin technology, this modeling process enhances the reliability, performance, and safety of UAVs, empowering informed decision-making and operational optimization in the realm of unmanned aerial systems. Secondly, to incorporate the learning capacity in low-level control, a novel iterative learning sliding mode controller is developed for trajectory tracking of quadrotor UAVs under model uncertainties and external disturbances. Integrated in the outer loop of a controlled system, this discrete-time control design eliminates the need for prior knowledge of disturbance bounds. The application of this design to the attitude control of a 3DR Solo UAV, alongside a built-in PID controller, is validated through simulations and real-time experiments, demonstrating its superiority over existing techniques. Thirdly, for high-level control of UAVs, innovative cooperative path planning algorithms are designed, addressing complex environments through the formulation of game theory and particle swarm optimization (PSO). Algorithms are developed for the stag-hunt and Nash-Stackelberg games. Next, the Fermat-Weber location is integrated into PSO algorithms aiming at reaching global optimal solutions. Experimental tests on a group of three UAVs confirm the advantages of the proposed approach in practical applications. Finally, the thesis presents a comprehensive approach for surface inspection of built infrastructure, with a focus on monorail bridges and building fa¸cade, utilizing UAVs and the digital twin technology. The approach includes an autonomous UAV-based inspection system with four key components: UAV dynamics, tracking control, path planning, and task execution. The methodology is validated through simulations and real-world experiments, affirming its effectiveness in authentic scenarios. The thesis results underscore the potential of the proposed approaches in addressing the evolving intelligence in various applications of UAVs.

Author dc:creator

• Nguyen, Van Lanh

Statement dc:rights

• info:eu-repo/semantics/openAccess
• The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
• © 2024 Van Lanh Nguyen
• au.edu.uts.lib/cph

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/10453/181062

OAI identifier oai:identifier

oai:opus.lib.uts.edu.au:10453/181062