Model-Scale Evaluation of Autonomous Ship Landing Guidance and Control Modes for Rotorcraft
Open Access
- Author:
- Hendrick, Christopher
- Graduate Program:
- Aerospace Engineering (PHD)
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 22, 2024
- Committee Members:
- Amy Pritchett, Program Head/Chair
Jacob Langelaan, Major Field Member
Joseph Horn, Chair & Dissertation Advisor
Chris Rahn, Outside Unit & Field Member
Eric Johnson, Major Field Member - Keywords:
- VTOL
Rotorcraft
Ship Landing
Autonomous
Froude Scaling
Model-Scale
UAV
Guidance
Control - Abstract:
- Ship landing in high sea states is a challenge for both manned and unmanned rotorcraft. A system that provides reliable autonomous recovery of ship-based rotorcraft could reduce mishap rates and reduce costs associated with training and certification testing. Such a system might also increase operational capability by allowing operations in more severe wave and wind conditions. These potential benefits have motivated a significant amount of public domain research on autonomous ship landing algorithms for rotorcraft. Existing works have primarily been simulation based, however, and the rigorous experimental evaluation of advanced landing algorithms is lacking in the public-domain literature. This is due in part to the inaccessibility of full-scale testing. Model-scale testing, on the other hand, offers a more accessible test bed for vetting autonomous landing solutions and has therefore been utilized in all existing openly available experimental studies. These studies have not considered the scaling of either the closed-loop aircraft dynamics or ship motions, however, meaning key aspects of the full-scale landing scenario may not be realistically represented at model-scale. The objective of this research was to develop a methodology for performing dynamically scaled autonomous ship landing experiments, and then to use the proposed scaling method to perform a rigorous experimental analysis of advanced autonomous landing guidance and control modes. Toward this end, Froude scaling is proposed for relating aircraft closed-loop dynamics and ship motions across test scales and the validity of this method is then analyzed. Two representative landing guidance algorithms were then developed and experimentally evaluated using the proposed scaling methodology. The first is an advanced landing strategy that uses quadratic programming (QP) optimization to plan the landing path to a forecasted deck state. The second is a simpler ``baseline’’ guidance method that tracks deck motions while closing the distance between the aircraft and deck at a constant rate. Both guidance algorithms command position and heading to an explicit model following (EMF) control law. The guidance algorithms were first evaluated in experiments conducted in the Maneuvering and Seakeeping Basin (MASK) located at the Naval Surface Warfare Center Carderock Division. During these experiments, control law parameters were modified to impose artificial constraints on the maneuverability of the aircraft, providing insight into how well both guidance methods can cope with a less agile airframe. The results showed that the predictive landing strategy allowed for more direct landing paths to be planned when compared to the baseline algorithm and can also allow for landings to be performed with lower control bandwidth. The baseline guidance algorithm, on the other hand, proved to be both simple and reliable when the UAV was in high bandwidth configurations, but may not be feasible for aircraft with limited control authority that must land in moderate to high sea states. The experimental setup in the MASK did not include aerodynamic disturbances. Additional flight tests were therefore performed in the Penn State Indoor Flight Facility to determine if the QP guidance algorithm offers any advantage over the baseline method when operating in the presence of a significant aerodynamic disturbance. The experimental results obtained at Penn State did not indicate that the QP algorithm provides a definitive advantage over the baseline algorithm in terms of matching deck position, velocity, and attitude at landing while operating in a gusty environment. The QP algorithm did allow for landings with a shorter duration, however, resulting in lower total control usage due to less time operating in aerodynamic disturbances.