Planificación de movimiento para mini dron con MPC.
Loading...
Identifiers
Publication date
Reading date
Authors
Collaborators
Tutors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Share
Department/Institute
Abstract
En los últimos años, los drones han ganado protagonismo en múltiples sectores por su capacidad de operar de forma autónoma y versátil. No obstante, lograr que se desplacen con precisión y seguridad sin intervención humana sigue siendo un reto.
Este Trabajo Fin de Máster plantea una solución basada en técnicas avanzadas de control, para ello, se ha desarrollado un modelo dinámico del quadrotor y se ha implementado un controlador MPC en un entorno de simulación basado en ROS, Gazebo y CasADI. Se llevaron a cabo múltiples experimentos para ajustar los parámetros críticos del controlador, como el horizonte de predicción y el intervalo de tiempo, identificando las combinaciones que ofrecen un mejor rendimiento. El siguiente paso lógico será trasladar los avances obtenidos en simulación a pruebas con un dron físico, permitiendo así validar su eficacia en condiciones reales y continuar explorando mejoras.
In recent years, drones have gained prominence across multiple sectors due to their ability to operate autonomously and flexibly. However, achieving accurate and safe navigation without human intervention remains a challenge. This Master's Thesis proposes a solution based on advanced control techniques. A dynamic model of the quadrotor was developed, and an MPC controller was implemented in a simulation environment using ROS, Gazebo, and CasADI. Multiple experiments were conducted to fine-tune the controller’s critical parameters, such as the prediction horizon and time step, identifying the combinations that offer the best performance. The next logical step will be to transfer the progress achieved in simulation to tests with a physical drone, thereby validating its effectiveness under real-world conditions and continuing to explore improvements.
In recent years, drones have gained prominence across multiple sectors due to their ability to operate autonomously and flexibly. However, achieving accurate and safe navigation without human intervention remains a challenge. This Master's Thesis proposes a solution based on advanced control techniques. A dynamic model of the quadrotor was developed, and an MPC controller was implemented in a simulation environment using ROS, Gazebo, and CasADI. Multiple experiments were conducted to fine-tune the controller’s critical parameters, such as the prediction horizon and time step, identifying the combinations that offer the best performance. The next logical step will be to transfer the progress achieved in simulation to tests with a physical drone, thereby validating its effectiveness under real-world conditions and continuing to explore improvements.
Description
Bibliographic citation
Collections
Endorsement
Review
Supplemented By
Referenced by
Creative Commons license
Except where otherwised noted, this item's license is described as Attribution-NonCommercial-NoDerivatives 4.0 International










