Kelly J. Hayhurst

Unmanned Aircraft Hazards and their Implications for Regulation

Use of unmanned aircraft systems (UASs) has been characterized as the next great step forward in the evolution of civil aviation. Indeed, UASs are in limited civil use in the United States today, and many believe that the time is rapidly approaching when they will move into the commercial marketplace, too. To make this a reality, a number of challenges must be overcome to develop the necessary regulatory framework for assuring safe operation of this special class of aircraft. This paper discusses some of what must be done to establish that framework. In particular, we examine hazards specific to the design, operation, and flight crew of UASs, and discuss implications of these hazards for existing policy and guidance. Understanding unique characteristics of UASs that pose new hazards is essential to developing a cogent argument, and the corresponding regulatory framework, for safely integrating these aircraft into civil airspace

A factoring approach for the stochastic shortest path problem

This paper studies the problem of determining the exact distribution of shortest path length in directed stochastic networks. Our approach is based on the concept of structural factoring, in which a stochastic network is decomposed into an equivalent set of smaller, generally less complex subnetworks. Several network constructs are identified and exploited to reduce significantly the computational effort required to solve a problem relative to complete enumeration. This algorithm can be applied to two important classes of stochastic network problems: determining the critical path length distribution for acyclic networks and the two-terminal reliability for probabilistic networks. Computational experience with the algorithm has been encouraging and has allowed the exact solution of networks previously analyzed only by approximation techniques.

A case study for assured containment

While incremental steps are being taken to integrate unmanned aircraft systems (UAS) into the various national airspace systems, much work remains to establish appropriate regulatory infrastructure that allows UAS larger than 55 lb to operate for commerce or hire. The magnitude of that effort is compounded by the wide-ranging variety of UAS types and possible applications, as well as the diversity in quality and provenance of UAS components. The FAA has suggested developing design standards tailored to specific applications and operating environments as an approach to facilitate integration and safe operation of some UAS. This paper introduces a case study to investigate design standards for a midsize unmanned rotorcraft operating in a rural environment. A key aspect of this study is the concept of using a certifiable containment system, different from a conventional geofencing application, to ensure that the unmanned aircraft does not escape its intended operational area. The proposed assured containment system is expected to reduce the effort needed to regulate some UAS that could not currently meet rigorous aircraft design standards and fall outside of the parameters for operation outlined in the proposed small UAS rule. This paper discusses how assured containment may be a useful approach to limiting risk and reducing an otherwise prohibitive certification burden to enable UAS operations in confined areas. The case study examines the potential effect the assured containment approach might have on airworthiness certification requirements.

SAFEGUARD: An assured safety net technology for UAS

As demands increase to use unmanned aircraft systems (UAS) for a broad spectrum of commercial applications, regulatory authorities are examining how to safely integrate them without loss of safety or major disruption to existing airspace operations. This work addresses the development of the Safeguard system as an assured safety net technology for UAS. The Safeguard system monitors and enforces conformance to a set of rules defined prior to flight (e.g., geospatial stay-out or stay-in regions, speed limits, altitude limits). Safeguard operates independently of the UAS autopilot and is strategically designed in a way that can be realized by a small set of verifiable functions to simplify compliance with regulatory standards for commercial aircraft. A framework is described that decouples the system from any other devices on the UAS as well as introduces complementary positioning source(s) for applications that require integrity and availability beyond what the Global Positioning System (GPS) can provide. Additionally, the high level logic embedded within the software is presented, as well as the steps being taken toward verification and validation (V&V) of proper functionality. Next, an initial prototype implementation of the described system is disclosed. Lastly, future work including development, testing, and system V&V is summarized.

Considerations of Unmanned Aircraft Classification for Civil Airworthiness Standards

The use of unmanned aircraft in the National Airspace System (NAS) has been characterized as the next great step forward in the evolution of civil aviation. Although use of unmanned aircraft systems (UAS) in military and public service operations is proliferating, civil use of UAS remains limited in the United States today. This report focuses on one particular regulatory challenge: classifying UAS to assign airworthiness standards. Classification is useful for ensuring that meaningful differences in design are accommodated by certification to different standards, and that aircraft with similar risk profiles are held to similar standards. This paper provides observations related to how the current regulations for classifying manned aircraft, based on dimensions of aircraft class and operational aircraft categories, could apply to UAS. This report finds that existing aircraft classes are well aligned with the types of UAS that currently exist; however, the operational categories are more difficult to align to proposed UAS use in the NAS. Specifically, the factors used to group manned aircraft into similar risk profiles do not necessarily capture all relevant UAS risks. UAS classification is investigated through gathering approaches to classification from a broad spectrum of organizations, and then identifying and evaluating the classification factors from these approaches. This initial investigation concludes that factors in addition to those currently used today to group manned aircraft for the purpose of assigning airworthiness standards will be needed to adequately capture risks associated with UAS and their operations.

SAFEGUARD: Progress and test results for a reliable independent on-board safety net for UAS

As demands increase to use unmanned aircraft systems (UAS) for a broad spectrum of commercial applications, regulatory authorities are examining how to safely integrate them without compromising safety or disrupting traditional airspace operations. For small UAS, several operational rules have been established; e.g., do not operate beyond visual line-of-sight, do not fly within live miles of a commercial airport, do not fly above 400 ft above ground level. Enforcing these rules is challenging for UAS, as evidenced by the number of incident reports received by the Federal Aviation Administration (FAA). This paper reviews the development of an onboard system — Safeguard — designed to monitor and enforce conformance to a set of operational rules defined prior to flight (e.g., geospatial stay-out or stay-in regions, speed limits, and altitude constraints). Unlike typical geofencing or geo-limitation functions, Safeguard operates independently of the off-the-shelf UAS autopilot and is designed in a way that can be realized by a small set of verifiable functions to simplify compliance with existing standards for safety-critical systems (e.g. for spacecraft and manned commercial transportation aircraft systems). A framework is described that decouples the system from any other devices on the UAS as well as introduces complementary positioning source(s) for applications that require integrity and availability beyond what can be provided by the Global Positioning System (GPS). This paper summarizes the progress and test results for Safeguard research and development since presentation of the design concept at the 35th DASC (2016). Significant accomplishments include completion of software verification and validation in accordance with NASA standards for spacecraft systems (to Class B), development of improved hardware prototypes, development of a simulation platform that allows for hardware-in-the-loop testing and fast-time Monte Carlo evaluations, and flight testing on multiple air vehicles. Integration testing with NASAs UAS Traffic Management (UTM) service-oriented architecture was also demonstrated.

Some impacts of risk-centric certification requirements for UAS

This paper discusses results from a recent study that investigates certification requirements for an unmanned rotorcraft performing agricultural application operations. The process of determining appropriate requirements using a risk-centric approach revealed a number of challenges that could impact larger UAS standardization efforts. Fundamental challenges include selecting the correct level of abstraction for requirements to permit design flexibility, transforming human-centric operational requirements to aircraft airworthiness requirements, and assessing all hazards associated with the operation.

Application of industry-standard guidelines for the validation of avionics support

The application of industry standards to the development of avionics software is discussed, focusing on verification and validation activities. It is pointed out that the procedures that guide the avionics software development and testing process are under increased scrutiny. The Radio Technical Commission for Aeronautics DO-178A guidelines, Software Considerations in Airborne Systems and Equipment Certification, are used by the US Federal Aviation Administration for certifying avionics software. To investigate the effectiveness of the DO-178A guidelines for improving the quality of avionics software, guidance and control software (GCS) is being developed according to the DO-178A development method. It is noted that, due to the extent of the data collection and configuration management procedures, any phase in the life cycle of a GCS implementation can be reconstructed. Hence, a fundamental development and testing platform has been established that is suitable for investigating the adequacy of various software development processes. In particular, the overall effectiveness and efficiency of the development method recommended by the DO-178A guidelines are being closely examined.<<ETX>>

Testing of reliability-Analysis tools

An outline is presented of issues raised in verifying the accuracy of reliability analysis tools. State-of-the-art reliability analysis tools implement various decomposition, aggregation, and estimation techniques to compute the reliability of a diversity of complex fault-tolerant computer systems. However, no formal methodology has been formulated for validating the reliability estimates produced by these tools. The author presents three states of testing that can be performed on most reliability analysis tools to effectively increase confidence in a tool. These testing stages were applied to the SURE (semi-Markov unreliability range evaluator) reliability analysis tool, and the results of the testing are discussed.<<ETX>>