Quadcopter control exposed to ground, ceiling, wall and wind disturbances
Author
Pedersen, Tinus Stengaard
Term
4. term
Education
Publication year
2021
Submitted on
2021-11-15
Pages
74
Abstract
Droner bruges i mange områder til at gøre komplekse opgaver enklere. For at bruge dem sikkert og effektivt, især i industrien, kræves robust styring, der kan håndtere forstyrrelser fra omgivelserne. Denne afhandling undersøger, hvordan typiske forstyrrelser påvirker en quadcopter: målestøj, vind samt ændringer i luftstrøm tæt på gulv, loft og vægge. Modellerne for disse forstyrrelser er hentet fra tidligere studier. Tre styringsstrategier sammenlignes for at belyse forskellen mellem en simpel og en mere avanceret tilgang: en lineær PD-regulator og to sliding mode-metoder (SMC og ISMC). PD justerer efter afvigelsen fra målet og hvor hurtigt den ændrer sig, mens sliding mode-styring tvinger systemet til at følge en robust regel, der bedre tåler forstyrrelser; ISMC tilføjer et integrerende led for yderligere robusthed. For at dæmpe virkningen af støj er der udviklet et udvidet Kalman-filter, som kombinerer støjfyldte sensordata for at give renere estimater. Testene viser, at sliding mode-styring er mere robust over for forstyrrelser, og at lineær styring var uanvendelig i flere tilfælde. Afhandlingen fremhæver derfor, at avanceret styring bør anvendes i udfordrende miljøer, når høj ydeevne er vigtig, og understreger betydningen af robust styring i industrielle droneanvendelser.
Drones are used in many fields to simplify complex tasks. To use them safely and effectively, especially in industry, we need robust control that can handle environmental disturbances. This thesis examines how common disturbances affect a quadcopter: measurement noise, wind, and airflow changes near the ground, ceiling, and walls. The disturbance models are based on earlier studies. Three control strategies are compared to highlight the difference between a simple and a more advanced approach: a linear PD controller and two sliding mode methods (SMC and ISMC). PD adjusts based on how far the drone is from its target and how fast that error changes, while sliding mode control forces the system to follow a robust rule that tolerates disturbances; ISMC adds an integral term for additional robustness. To reduce the impact of noise, an Extended Kalman filter was designed to combine noisy sensor measurements into cleaner estimates. Tests show that sliding mode control is more robust against disturbances, and that linear control was unusable in several cases. The thesis concludes that advanced control should be used in challenging environments when high performance is desired and highlights the importance of robust control for industrial drone applications.
[This summary has been rewritten with the help of AI based on the project's original abstract]
Keywords
Documents