Organizational Unit:
Daniel Guggenheim School of Aerospace Engineering
Daniel Guggenheim School of Aerospace Engineering
Permanent Link
Research Organization Registry ID
Description
Previous Names
Parent Organization
Parent Organization
Organizational Unit
Includes Organization(s)
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
Organizational Unit
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 4 of 4
-
ItemA Framework for Integrating Advanced Air Mobility Vehicle Development, Safety and Certification(Georgia Institute of Technology, 2022-04-28) Markov, AlexanderAs urbanization continues to grow world wide, cities are experiencing challenges dealing with the increases in pollution, congestion, and availability of public transportation. A new market in aviation, Advanced Air Mobility, has emerged to address these challenges by engineering novel aircraft that are all electric and meant to transport people within and between cities quickly and efficiently. The scale of this market and the associated operations means that vehicles will need to fly with increased autonomy. The lack of highly trained and skilled pilots, along with the increased work load for novel aircraft makes piloted aircraft infeasible at the scale intended or Advanced Air Mobility. While a variety of concepts have been created to meet the performance needs of such operations, the safety and certification requirements of these aircraft remain unclear. The paradigm shift from conventional aircraft to novel, highly integrated, and autonomous aircraft presents many challenges which motivate this work. An emphasis is placed on the safety assessment and the gaps between current regulations and the needs for Advanced Air Mobility. The research objective of this work is to develop a framework for the development and safety assessment of autonomous Advanced Air Mobility aircraft by first examining the existing methods, techniques, and regulations. In doing so, several gaps are identified pertaining to the hazard analysis, reliability analysis of Integrated Modular Avionics systems, and the inclusion of a Run-Time Assurance architecture for vehicle control. An improved hazard analysis approach is developed to capture functional failures as well as systematic areas that can lead to unsafe system behavior. The Systems-Theoretic Process Analysis is supplemented to the Continuous Functional Hazard Assessment so that system behavior and component interactions can be captured. Unsafe system and component actions are identified and used to develop loss scenarios which provide context to the specific conditions that lead to loss of critical vehicle functionality. This information is traced back to identified hazards and used to establish constraints to mitigate unsafe behavior. The Functional Hazard Assessment is then applied to applicable scenarios to provide severity and risk information so that quantitative metrics can be used in additional to qualitative ones. The improved approach develops requirements and determines component and system constraints so that requirements can be refined. It also develops a control structure of the system and assigns traceable items at each step to track how unsafe actions, losses, hazards, and constraints are linked. To improve the reliability modeling of complex modular avionics systems utilizing Multi-Core Processing, a Dynamic Bayesian Network modeling method is developed. This method first utilizes the existing methods defined in ARP 4761 for reliability analysis, namely the Fault Tree Analysis. A mapping is identified for converting fault trees to Bayesian networks, before a Dynamic Bayesian Network is developed by defining how component reliability changes with time. The capability to model reliability of these kinds of systems overtime alone is useful for developing and evaluating maintenance schedules. Additionally, it can handle degradable and repairable components and has the capability to infer failure probabilities using observed evidence. This is useful for identifying weak areas of the system that may be the most likely to cause an overall system failure. A secondary capability is the modeling of uncertainty and the reliability impacts of Multi-Core Processing factors. Subject Matter Expert input and test data can be used to develop conditional dependencies between factors like Worst-Case Execution time, complexity, and partitioning of multi-core systems and their impact on the reliability of the Real-Time Operating System. The added safety challenges of interference and system complexity can be modeled earlier in the design process and can quickly be updated as more information becomes available. Finally, the safe inclusion of autonomy is addressed. To do so, a Simplex architecture is chosen for the development and testing of complex controllers. These controllers are non0deterministic in nature and would otherwise not be certifiable as a result. The Simplex architecture uses an assured back up controller that is triggered when a monitor senses that some predefined safety threshold is breached and gives control back once the system is back to nominal operations. This architecture enables the use of complex control and functionality while also enabling the overall system to be certified. A model predictive control algorithm is developed using a recursive neural network and a receding horizon control scheme that allows a simple system to be controlled quickly and accurately. A PID controller is used as the assured back up controller and the monitoring and triggering capability is demonstrated. The architecture successfully triggers the back up when a threshold is exceeded and hands control back over to the complex controller when the system is brought back to nominal conditions. The main contribution of this dissertation is the development of a modified development assurance and safety management framework that is applicable to Advanced Air Mobility aircraft. The modifications made are specifically targeted at the challenges of applying the existing framework to novel, integrated, complex, and autonomous aircraft. This supports the objective of this research and provides guidance for how existing well understood and trusted methods can be modified for novel applications.
-
ItemShip and Naval Technology Trades-Offs for Science And Technology Investment Purposes(Georgia Institute of Technology, 2022-01-14) Gradini, RaffaeleLong-term naval planning has always been a challenge, but in recent years the difficulty has increased. The degradation of the security environment is leading toward a more volatile, uncertain, complex, and ambiguous world, heavily affecting the quality of predictions needed in long-term defense technology investments. This work tackles the problem from the perspective of the maritime domain, with a new approach stemming from the state-of-the-art in the defense investment field. Moving away from classic methodologies that rely on well-defined assumptions, it is possible to find investment processes that are broad enough, yet concrete, to support decision making in naval technology trades for science and technology purposes. In fulfilling this objective, this work is divided in two main areas: identifying technological gaps in the security scenario and providing robust technology investment strategies to cover those gaps. The core of the first part is the capability of decomposing maritime assets using modern taxonomies, to map the impact of different technologies on ships. Once technologies are mapped, they can be traded inside assets, and assets inside fleets to quantitatively evaluate the overall fleet robustness. The first deliverable achieved through this process is called Vulnerable Scenarios, a list of possible conflict scenarios in which a tested fleet would consistently fail. The second deliverable is called Robust Strategies and is made of different technological investments to allow the studied fleet in succeeding the discovered Vulnerable Scenario. To find the first deliverable a large set of scenarios were simulated. The results of this simulation were analyzed using the Patient Rule Induction Method to isolate, among the large set of relevant cases, a subgroup of Vulnerable Scenarios. These were identified by highlight commonalities on shared parameters and variables. Once the Vulnerable Scenarios were discovered, an ad-hoc adaptive response system using a “signpost and trigger” mechanism was used to identify different technologies on the ships studied that could enhance the overall robustness of the fleet. In identifying these technologies, the adaptive system was supported by different taxonomies in performing the different technological trades that allowed the algorithm to find Robust technology Strategies. The methodology was completed by a ranking system that was designed to firstly check all the Robust Strategies in all the scenarios of interest, and then to compare them against ranking metrics defined by decision makers. To test the created methodology, several experiments were conducted across two use cases. The first use case, which involved an anti-submarine warfare (ASW) mission, was used to demonstrate the individual pieces employed in the creation of the methodology. The second use case, involving a large operation made of several tasks, was used to test the overall methodology as one. Both use cases were designed on the same original scenario created in collaboration with former generals and admirals of the US Air Force and the Italian Navy. The primary results of this experiments show that once Vulnerable Scenarios are discovered, it is possible to employ an iterative algorithm that recursively infuse new technologies into the fleet. This process is repeated until Robust Technology Strategies that can support the fleet are selected. The missions designed demonstrated the presence of gaps which had to be covered via technology investment showing how planners will have to account for new technologies to be able to succeed in future challenges. The methodology created in this thesis provided an innovative way of enhancing the screening of maritime scenarios, reducing the leading time for investment decisions on naval technologies. In conclusion, the work done in this thesis helps in advancing the state of the art of methodologies used by planners when looking for Vulnerable Scenarios and for new technologies to invest on. Therefore, this thesis demonstrates that by employing the proposed methodology, Vulnerable Scenarios and relevant technologies can be identified in less time than by employing current methods. These efforts will support planners and decision makers in reacting faster to new emerging threats in unforeseen naval scenarios and, will enable them to identify in a rapid fashion in which areas more investments are needed.
-
ItemA DATA-DRIVEN METHODOLOGY TO ANALYZE AIR TRAFFIC MANAGEMENT SYSTEM OPERATIONS WITHIN THE TERMINAL AIRSPACE(Georgia Institute of Technology, 2021-12-10) Corrado, Samantha JaneAir Traffic Management (ATM) systems are the systems responsible for managing the operations of all aircraft within an airspace. In the past two decades, global modernization efforts have been underway to increase ATM system capacity and efficiency, while maintaining safety. Gaining a comprehensive understanding of both flight-level and airspace-level operations enables ATM system operators, planners, and decision-makers to make better-informed and more robust decisions related to the implementation of future operational concepts. The increased availability of operational data, including widely-accessible ADS-B trajectory data, and advances in modern machine learning techniques provide the basis for offline data-driven methods to be applied to analyze ATM system operations. Further, analysis of ATM system operations of arriving aircraft within the terminal airspace has the highest potential to impact safety, capacity, and efficiency levels due to the highest rate of accidents and incidents occurring during the arrival flight phases. Therefore, motivating this research is the question of how offline data-driven methods may be applied to ADS-B trajectory data to analyze ATM system operations at both the flight and airspace levels for arriving aircraft within the terminal airspace to extract novel insights relevant to ATM system operators, planners, and decision-makers. An offline data-driven methodology to analyze ATM system operations is proposed involving the following three steps: (i) Air Traffic Flow Identification, (ii) Anomaly Detection, and (iii) Airspace-Level Analysis. The proposed methodology is implemented considering ADS-B trajectory data that was extracted, cleaned, processed, and augmented for aircraft arriving at San Francisco International Airport (KSFO) during the full year of 2019 as well as the corresponding extracted and processed ASOS weather data. The Air Traffic Flow Identification step contributes a method to more reliably identify air traffic flows for arriving aircraft trajectories through a novel implementation of the HDBSCAN clustering algorithm with a weighted Euclidean distance function. The Anomaly Detection step contributes the novel distinction between spatial and energy anomalies in ADS-B trajectory data and provides key insights into the relationship between the two types of anomalies. Spatial anomalies are detected leveraging the aforementioned air traffic flow identification method, whereas energy anomalies are detected leveraging the DBSCAN clustering algorithm. Finally, the Airspace-Level Analysis step contributes a novel method to identify operational patterns and characterize operational states of aircraft arriving within the terminal airspace during specified time intervals leveraging the UMAP dimensionality reduction technique and DBSCAN clustering algorithm. Additionally, the ability to predict, in advance, a time interval’s operational pattern using metrics derived from the ASOS weather data as input and training a gradient-boosted decision tree (XGBoost) algorithm is provided.
-
ItemOptimal aeroelastic trim for rotorcraft with constrained, non-unique trim solutions(Georgia Institute of Technology, 2008-02-15) Schank, Troy C.New rotorcraft configurations are emerging, such as the optimal speed helicopter and slowed-rotor compound helicopter which, due to variable rotor speed and redundant lifting components, have non-unique trim solution spaces. The combination of controls and rotor speed that produce the best steady-flight condition is sought among all the possible solutions. This work develops the concept of optimal rotorcraft trim and explores its application to advanced rotorcraft configurations with non-unique, constrained trim solutions. The optimal trim work is based on the nonlinear programming method of the generalized reduced gradient (GRG) and is integrated into a multi-body, comprehensive aeroelastic rotorcraft code. In addition to the concept of optimal trim, two further developments are presented that allow the extension of optimal trim to rotorcraft with rotors that operate over a wide range of rotor speeds. The first is the concept of variable rotor speed trim with special application to rotors operating in steady autorotation. The technique developed herein treats rotor speed as a trim variable and uses a Newton-Raphson iterative method to drive the rotor speed to zero average torque simultaneously with other dependent trim variables. The second additional contribution of this thesis is a novel way to rapidly approximate elastic rotor blade stresses and strains in the aeroelastic trim analysis for structural constraints. For rotors that operate over large angular velocity ranges, rotor resonance and increased flapping conditions are encountered that can drive the maximum cross-sectional stress and strain to levels beyond endurance limits; such conditions must be avoided. The method developed herein captures the maximum cross-sectional stress/strain based on the trained response of an artificial neural network (ANN) surrogate as a function of 1-D beam forces and moments. The stresses/strains are computed simultaneously with the optimal trim and are used as constraints in the optimal trim solution. Finally, an optimal trim analysis is applied to a high-speed compound gyroplane configuration, which has two distinct rotor speed control methods, with the purpose of maximizing the vehicle cruise efficiency while maintaining rotor blade strain below endurance limit values.