Browsing by Subject "Robust control"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Finite Horizon Robustness with Applications to Missile Engagements
    (2021-12) Buch, Jyot
    This dissertation advances theoretical and computational tools for robustness analysis and synthesis on a finite time horizon. A motivation for this work is the reliable assessment of missile interceptor system performance, which will also enable the robust design of such systems. Typical performance metrics are infinite-horizon in nature, centered on stability, and rely on frequency-domain concepts, such as gain/phase margins. Such metrics can be inadequate for systems operating on finite-time horizons, as in many launch scenarios. Instead, this thesis focuses on time-domain metrics, e.g., bounds on the system's state at the final time of the horizon, while considering the impact of disturbances, model uncertainty/variability, and initial conditions. The proposed approach is to numerically linearize the dynamics along the trajectory to obtain a Linear Time-Varying (LTV) system. The analysis or synthesis is then performed on a linearized system that captures the dynamics up to the first-order perturbations around the nominal trajectory. This sacrifices some accuracy over the original nonlinear model but enables the use of linear system tools. The proposed worst-case LTV analysis also provides specific bad disturbances and uncertain parameters that can be further studied in high-fidelity nonlinear simulation.
  • Thumbnail Image
    Item
    Modeling, Robust Control, and Experimental Validation of a Supercavitating Vehicle
    (2015-06) Escobar Sanabria, David
    This dissertation considers the mathematical modeling, control under uncertainty, and experimental validation of an underwater supercavitating vehicle. By traveling inside a gas cavity, a supercavitating vehicle reduces hydrodynamic drag, increases speed, and minimizes power consumption. The attainable speed and power efficiency make these vehicles attractive for undersea exploration, high-speed transportation, and defense. However, the benefits of traveling inside a cavity come with difficulties in controlling the vehicle dynamics. The main challenge is the nonlinear force that arises when the back-end of the vehicle pierces the cavity. This force, referred to as planing, leads to oscillatory motion and instability. Control technologies that are robust to planing and suited for practical implementation need to be developed. To enable these technologies, a low-order vehicle model that accounts for inaccuracy in the characterization of planing is required. Additionally, an experimental method to evaluate possible pitfalls in the models and controllers is necessary before undersea testing. The major contribution of this dissertation is a unified framework for mathematical modeling, robust control synthesis, and experimental validation of a supercavitating vehicle. First, we introduce affordable experimental methods for mathematical modeling and controller testing under planing and realistic flow conditions. Then, using experimental observations and physical principles, we create a low-order nonlinear model of the longitudinal vehicle motion. This model quantifies the planing uncertainty and is suitable for robust controller synthesis. Next, based on the vehicle model, we develop automated tools for synthesizing controllers that deliver a certificate of performance in the face of nonlinear and uncertain planing forces. We demonstrate theoretically and experimentally that the proposed controllers ensure higher performance when the uncertain planing dynamics are considered. Finally, we discuss future directions in supercavitating vehicle control.