Recoverability envelope analysis of nonlinear control laws for agile maneuvering aircraft

This theses proposes a novel framework to examine the effectiveness of controllers, which are designed for agile maneuvering aircraft upset recovery. The introduced framework incorporates two main assessment criteria. First, the angular rate envelope extent of the controller is determined, which provides a measure of functionality region of the controller. Second, time histograms are obtained in order to understand how fast does the controller recover the aircraft if the controller can recover the aircraft. Such recovery time analysis also lead this theses to cover a novel methodology which yields the recovery of an agile maneuvering aircraft from upset conditions as swiftly as possible. In this work, a linear quadratic regulator, a nonlinear dynamic inversion (NDI) based angular rate controller, double loop NDI controller, and another NDI based switching controller are considered for comparison. The latter controller presented in this theses is a new strategy for upset recovery of an agile maneuvering aircraft. As it is mentioned earlier, this theses introduces a novel methodology in order to recover an agile maneuvering aircraft from upset conditions as swiftly as possible. The current methodology based on an angular rate grid, which consists of angular velocity nodes. The aircraft is said to start its motion from an initial condition while the initial angular velocity corresponds to a certain node within the aforementioned angular rate grid. A nonlinear dynamic inversion (NDI) based controller is employed to reach transition angular rates, which are the desired rates before the recovery initiates. Once the desired transition node is reached, the controller produces necessary input signals to recover the aircraft. The time elapsed throughout the whole recovery process is recorded. The same sequence of control actions are repeated for each initial and transition nodes of the grid. In this way, a recovery time graph is obtained. The time graph is then used to determine the reference angular rate trajectory, which provides the minimum recovery time for the given rate envelope. Simulations are carried out by employing a nonlinear high fidelity F-16 model. Simulation results show that the switching NDI controller has the superior recovery performance in terms of angular rate envelope extent. Moreover it is revealed that there exist such angular rate trajectories results with faster recovery. Therefore, the aforementioned results point out non-intuitive solutions. That is, for certain initial states it is not suitable to directly regulate the angular rates when recovery time is considered. Rather, for some cases, it is more convenient to increase the roll rate first, and then regulate the angular rates. That being said, the recovery methodology proposed in this theses yields faster recovery.