| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 10.65 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Technological advances have greatly expanded the use of robotics in areas such as marine engineering and scientific research. Robots have been fundamental in solving complex problems in hazardous environments, improving safety and efficiency by minimising human intervention. The development of autonomous or unmanned vehicles is particularly noteworthy. These vehicles often cooperate, requiring improved methods for maintaining optimal operational relationships. This dissertation addresses the challenge of optimising the geometry, stabilization and control of a parallel mechanism, integrated into an autonomous surface vehicle, for the landing of an unmanned aerial vehicle. To ensure a safe landing, the platform must be horizontally stable,
even under harsh offshore wave conditions. The present work focuses on enhancing the mechanism’s structural stability and wave-motion compensation capability. Firstly, a detailed study of the mechanism’s design is presented, with structural improvements validated through finite element analysis. This study is followed by a methodology for short-term wave forecasting, using a Butterworth low-pass filter and an auto-regressive prediction model. A constrained model-based predictive control method is then used to optimise the trajectory of the landing platform. Finally, the presented methods and modifications are implemented, and
validated in a simulated environment. The results of the finite element analysis demonstrate that the design improvements have significantly increased the structural stability of the parallel mechanism, which was verified in the real structure. Additionally, the on-line implementation of the controller exhibited a positive performance in the wave-motion compensation task, maintaining the landing platform in a nearly horizontal position at almost every analysed time instance.
Descrição
Palavras-chave
Autonomous surface vehicle Parallel mechanism Finite element analysis Butterworth low-pass filter Auto-regressive prediction model Model predictive control