Modelling And Control System Design For Autonomous Airship

In recent years, autonomous airships have become an intense research area all over the world for their emerging applications such as communication platform, surveillance, advertising, monitoring, inspection, exploration, and so on. This thesis studies unmanned airship modeling and control system design problems. The innovative work lies in following aspects:Modeling autonomous airship. As the basis of designing control system, the model of airship is needed. Airship model has two types of description in literatures. The first description of model is revised by considering vector thrust and the second description of model is developed using energy analysis and quasi-Lagrangian equation. Two types of nonlinear model are presented as the six-degree-of-freedom dynamic equations. Then the corresponding relationship of them is analyzed.Defining a family of error functions including position and attitude error, velocity error of airship. These error definitions are not the conventional differences between current values and desired values. A novel mehod is used to educe them. One more important point is that these errors have specific physical meanings. The model of error system is also established by differentiating the error equations along the state trajectory of the original system.Designing a nonlinear state feedback controller for airship with neutral buoyancy. Based on above error definition, a nonlinear state feedback controller is designed for airship with planar motion. First the nonlinear control law is given using Lyapunov method to stabilize the error system. Then the closed-loop system is proven to be uniformly asymptotically stable at the origin by applying Matrosov theorem. So system stabilization and trajectory tracking problems are solved unifiedly. Moreover, the controller is independent from airship's parameters such as mass, inertia. This ensures the robustness of controller with parameter uncertainty.Designing a nonlinear adaptive state feedback controller for airship with neutral buoyancy. Airship's model has parameter uncertainty because of gasbag-ballonets inflation-deflation, added inertial effects, viscous dissipation, wind disturbance and so on. The controller is often asked to be adaptive to ensure system performance. The nonlinear adaptive control law is given using Lyapunov method. Then the closed-loop system is proven to be locally uniformly asymptotically stable at the origin by applying Matrosov theorem.Designing H_∞controller based on convex polyhedron principles. The complexity of nonlinear controllers above brings much difficulty for engineering application though their control performances are perfect. The controller designed based on simple linearized model can't ensure the performance when system's states deviate from selected equilibrium point. A H_∞controller is designed to remedy this shortcoming. First airship's nonlinear model is equivalent to a convex polyhedron. Then a H_∞controller is designed using LMI conditions of robust controller for the convex polyhedron system.Predictive control system design for airship's planar motion. To overcome the shortcoming of H_∞controller, the control input saturation is fully taken into account during control system design. An optimal trajectory of"nominal system"is obtained by using a predictive controller. This trajecoty is also a feasible trajectory of"practical system". The H_∞control law is designed to make the system tracking the desired state trajectory. Simulation results verify the performance of controller.