Feedback Control of the National Airspace System

An Agent-Based Approach to Aircraft Conflict Resolution with Constraints

The current demand in the National Airspace System (NAS) is rapidly approaching the system capacity. This capacity limit causes significant congestion and delay in off-nominal conditions, such as inclement weather. It will also constrain future growth in the aviation sector. Transformation of the current system can relieve this constraint. Examples of such transformation include (i) flight schedule shifting to periods of low demand or alternate airports, (ii) air traffic control (ATC) productivity enhancements, (iii) new procedures and airspace redesigns, and (iv) improvements in weather nowcasting and forecasting. Planners are now investigating such NAS architecture strategies. This paper describes research and plans regarding our investigation of future NAS performance in inclement weather days. Our NAS simulation allows researchers and decision makers to hypothesize strategies in various weather conditions and traffic demand levels to quantify the combined impact on the NAS. The results of the simulation will guide decisions regarding the value of investments in weather forecasting technology and NAS procedures. Our simulation approach is to use (i) archived weather and traffic schedule data with (ii) weather and traffic forecasting models and (iii) tactical weather vectoring and stochastic traffic flow management capabilities.

This paper explores how applying a basic management science principle to air traffic management operations can lead to revised roles and responsibilities that allow operations to be handled by the most appropriate participants. Previous research that applied the principle to local departure and arrival management is reviewed; the application is then expanded to regional and national air traffic management. The paper examines the operational and engineering implications of the resulting concept, contrasted with current operations and tools, and identifies where additional research or development is needed. Examples are provided that illustrate the resulting activities at the various air traffic management facilities across a broad range of look-ahead timeframes. I. Introduction HE High-Density Area Departure and Arrival Management (HDDAM) concept was developed in the context of arrival and departure air traffic management in high-density airspace such as Metroplex areas. The management science principles underlying HDDAM appear to be applicable to National Airspace System (NAS) operations at any level, including all phases of planning and flight. This paper explores some operational and engineering implications of such an application to local, regional, and NAS-wide air traffic operations. As described elsewhere, 1-3 HDDAM redefines the decision making processes within the Air Route Traffic Control Center (ARTCC), the Terminal Radar Approach Control (TRACON), and multiple Air Traffic Control Towers (ATCTs) for high-density area departure/arrival traffic management. The decision making authority proposed for each facility reflects the information available to each decision maker and the impact of the decision on their operating environment. Specifically, each facility is responsible for defining the capacity of its air traffic management resources (such as airports, airspace, airways) and requesting resources for flights to use. As shown in Fig. 1 for a single Metroplex, the TRACON has a central role in these interactions, serving as a “broker” between the facilities with the resources and the facilities with requests for use of those resources that are not within their scope of control, while also accounting for the TRACON’s ability to deliver the flights to the resources in question. ATCTs are responsible for requesting resources for departing flights and the ARTCC for requesting resources for arriving flights, as they each possess the best information regarding the status of these flights. Because of the uncertainties present in predicting individual flight departure readiness and inbound air traffic arrival times, the TRACON does not assign specific departure and arrival slots to individual flights. Instead, the TRACON allocates slots to the airports (for departures) and ARTCC (for arrivals) so they can make flight-specific assignments as deemed best for their individual operating environments. This effectively decouples the decisions of the individual ATCTs and the ARTCC, giving each facility increased flexibility to handle situations dynamically as they develop. Air traffic management operations are most effective when the decision makers work as a team. The underlying principle of HDDAM is that this can be accomplished best when all facilities’ roles and responsibilities are based on the management science principle that allows the appropriate level of autonomy at each facility. The principle states that decisions should be assigned to those who are already performing other tasks with the relevant data—those who know the most about the factors affecting the decision—and will be most directly affected by the consequences of 1 Lead Multi-Discipline System Engineer, TFM System Engineering and Evolution, 7515 Colshire Drive, Mailstop

A number of Federal Aviation Administration (FAA) automation systems and developmental programs interoperate to support the day-to-day operations of the National Airspace System (NAS). The integration of New Entrants (NEs), such as Remotely Piloted Aircraft (RPA) 1 and Space Vehicle (SV) operations, into the NAS will require modification to the current state of automation. Examining the relevant operational, technical, and procedural implications of integrated NE operations, in addition to exploring the potential impacts of accelerated RPA usage on NAS dynamics, are critically important efforts necessary to understanding operational impacts to NAS systems and users. Our paper presents the potential impacts of RPA integration on two of these systems: En Route Automation Modernization (ERAM) and Traffic Flow Management System (TFMS). A significant component of the impact on NAS systems comes from flight plan and related data processing since all SVs and many RPAs will fly non-standard flight patterns (e.g., RPA loitering and grid search). The processes and details of the non-traditional flight routes, as well different RPA performance specifications, need to be considered as missions are overlaid on the national jet route structure. Of particular concern and interest are the changes in air traffic control (ATC) and air traffic management (ATM) systems and sub-systems needed to address air traffic controller workload and workload complexity concerns associated with NE activities. One potential accommodation approach, presented in this paper, is the use of Dynamic Airspace Configuration both for RPA routine operations and contingency events. Additionally, we present an RPA 1 In this paper, UAS and RPA are used interchangeably without intending any difference per se. It is noted that the U.S. DoD and ICAO use RPA while NASA and the FAA use UAS. demand forecast tool developed under a National Aeronautics and Space Administration (NASA) project. This tool will assist the FAA in meeting its objectives for accommodation of RPA by providing accurate demand forecast information.

The Airspace Concepts Evaluation System (ACES) is a distributed fast-time computer simulation used for design and trade-off studies of system level concepts for the National Airspace System (NAS). ACES employs multi-trajectory modeling in an agent-based software structure to describe the behavior and interactions of Traffic Flow Management, Air Traffic Control, Flight Deck, and Airline Operational Control functions and associated technologies. ACES modules include airport, terminal, en route and system-wide traffic management and controller agents, dispatcher agents, and flight agents; individual aircraft flight dynamics; and airport, airspace, meteorological, and related infrastructure components. ACES simulates gate-to-gate flight operations through the multi-sector, multiCenter, multi-airport NAS network. This paper focuses on the design and implementation of enhanced terminal area modeling capabilities in ACES Build 3. Previous builds of ACES have implemented a nodal model to describe runway system utilization at each airport with nominal terminal airspace transit times. The Build 3 capabilities provide improved modeling fidelity by accounting for individual runway operations and enabling the specification of additional detail in terminal airspace network operations. The Runway Modeling capability simulates individual aircraft operations on individual runways based on user-specified runway use configurations and runway interactions at an airport. The Simplified Terminal Airspace Network Modeling capability simulates irregular terminal airspace boundaries, variable terminal transit times and multiple airport operations in a terminal area. ACES Build 3 introduces capabilities to apply more complex terminal area modeling to selected airports while retaining the lesscomplex nodal model at other airports. ACES also supports variable complexity modeling by enabling users to define the time span over which Traffic Flow Management agents have accurate knowledge of upcoming airport operating conditions (versus inaccurate knowledge based on extrapolated forecasts) and to define the mode by which traffic flow restrictions are simulated (i.e., determined by runway system operating dynamics versus manually-defined static acceptance rates). These features support the modeling of decision support tools of variable levels of sophistication, enabling ACES users to assess the relative value of alternative advanced tool development efforts.

The National Aeronautics and Space Administration (NASA) is conducting far-term research into a proposed concept for gate-to-gate national airspace system (NAS) operations called Distributed Air/Ground Traffic Management (DAG-TM) in which flight deck crews, air traffic service providers and aeronautical operational control facilities use distributed decision-making. In this study we estimate the benefits of En Route Free Maneuvering and En Route Trajectory Negotiation in off-nominal conditions using empirical data. Specifically, the benefits we measure are the avoided disruption costs under inclement weather conditions due to the implementation of the two en route concept elements. The random nature of inclement weather makes the measurement of its impact on aviation a challenge. Since June 2003, the Department of Transportation's Airline Service Quality Performance (ASQP) data reports the major airlines' schedule/operation disruptions, in terms of flight cancellations, delays, and diversions, allocated to five cause categories, including extreme weather and non-extreme weather conditions. We extract the inclement weather- caused disruptions from the one-year period of July 2003 through June 2004 from the ASQP data. In developing the benefits pool, we recognize that these two concept elements' functionalities and capabilities could not eliminate all disruptions caused by weather conditions. We estimate the percentage of weather caused disruptions that can be helped by the two concept elements based on expert opinion. We estimate the NAS-wide benefits for year 2004 by extending the ASQP results to the non-ASQP carriers via linear extrapolation. We also extrapolate current NAS benefits to the projected benefits in year 2015, the target year of DAG-TM deployment, by developing and applying forecast parameters to year 2004 benefits based on the traffic forecast. Finally, we translate the operational performance benefits of reduced delay and avoided diversions/cancellations into economic benefits using FAA and other published cost factors. I. Introduction HE National Aeronautics and Space Administration (NASA) is conducting far-term research into a proposed concept for gate-to-gate national airspace system (NAS) operations called Distributed Air/Ground Traffic Management (DAG-TM). DAG-TM is based on distributed decision making between flight deck crews, air traffic service providers (ATSP) and aeronautical operational control (AOC) personnel. The goal of DAG-TM is to increase system capacity/throughput, enable user preferences, and provide greater flexibility and efficiency, while meeting air traffic management (ATM) requirements and maintaining system safety and user accessibility to the NAS. DAG-TM will be accomplished with a human-centered operational paradigm enabled by procedural and technological innovations. These innovations include automation aids, information sharing and Communication, Navigation, and Surveillance (CNS) / ATM technologies. The total DAG-TM concept is intended to address all user classes (commercial carriers, general aviation, etc.) with an emphasis towards ensuring access to airspace resources

The National Aeronautics and Space Administration (NASA) Advanced Air Transportation Technologies (AATT) project was charged with managing the research and development (R&D) of a portfolio of air traffic management (ATM) decision support tools (DSTs) and Distributed Air/Ground Traffic Management (DAG-TM) concept elements (CEs). These DSTs and CEs were designed to increase the capacity, efficiency, and flexibility of the National Airspace System (NAS) while maintaining the current level of safety of the NAS. This paper provides a case study of safety assessments in the context of an institutional R&D program, describing the safety assessment framework and methodology used by the AATT project to ensure that safety assessments met both NASA R&D objectives and FAA safety requirements. To illustrate the framework and methodology, this paper presents two AATT safety assessment case studies. The first uses results from the safety assessments of a DST to illustrate the AATT safety assessment framework and methodology. The second uses results from an interim safety assessment of a DAG-TM CE to illustrate how interim NASAinternal safety assessments were geared to provide constructive feedback to the R&D process. This paper underscores the idea that during the R&D phase, safety assessments must be conducted in a manner that provides the most constructive feedback/input to the R&D process.

In this paper, algorithm-generated airspace sector combinations are analyzed to determine if they are operationally acceptable. Discussions with practitioners identified three desirable characteristics for operational sector combinations. The first is that at each time during a weekly cycle, approximately the same number of sectors are operational. The second is that sector combinations be familiar, and the third is that sector combinations do not change frequently. Methods for visualizing and quantifying these characteristics are proposed. The methods are used to analyze historical sector combinations from the national airspace system and sector combinations generated by a sector-combining algorithm. The results were analyzed by researchers and practitioners. This analysis indicates that changes to the sector-combining algorithm may be required to ensure combinations that are acceptable for current operations. However, the identified characteristics may be less important in a mid-term timeframe (�2018), which is when a sector combining algorithm is being targeted for deployment. I. Introduction In current operations, supervisors in Air Route Traffic Control Centers (ARTCCs or centers) dynamically combine sectors when they are under-utilized (operating below their capacity). This is done primarily so that there are not more controller teams monitoring airspace than are necessary. There are many factors that impact supervisor decisions about when to combine and split sectors, such as air traffic levels, staff availability and capabilities, equipment availability, current and predicted weather, airport configurations, and air traffic demand management initiatives. Currently, these decisions are made based on supervisor experience and judgement. Some previous research has focused on predicting sector combinations and splits based on past supervisor decisions. 1 In Ref. 2, an algorithm to systematically suggest combinations of under-utilized airspace sectors has been proposed. Results suggest that this algorithm may utilize air traffic control resources more efficiently than they are currently utilized. However, feedback from practitioners regarding the sector combination algorithm in Ref. 2 suggests that the proposed sector combinations may not be operationally acceptable due to controller situational awareness and staff planning factors that are not explicitly considered in the algorithm. 3 In this research, the algorithm-suggested sector combinations are compared to historical sector combinations to determine if the algorithm produces operationally acceptable sector combinations. Three desirable characteristics of operational sector combinations specified by practitioners in Ref. 3 are investigated. Quantitative comparisons are designed to study each characteristic in the operational and algorithm-generated sector combinations. Further practitioner feedback on the results of these comparisons is reported.

A high-level state model of the national airspace system (NAS) has been developed and refined. That model provides guidance on the effect and interactions of sources of uncertainty in the NAS. The model also provides guidance on the possible effect of NextGen concepts and technologies on the NAS. The model is being coded as a high-level, abstract simulation that, when provided with specific uncertainty distributions, can also identify the emergent effect of those changes in the uncertainty distributions. The results of that analysis and simulation provide guidance for prioritizing the development and adoption of NextGen concepts and technologies for the NAS. I. Introduction he National Airspace System (NAS) is a highly complex, socio-technical system. At its lowest level, it consists of many thousands of entities, including aircraft and other sophisticated technology, passengers, pilots, and air traffic controllers. Modeling and simulation of such a system is difficult due to the large number of entities to be modeled, the infinite (or nearly so) number of degrees of freedom within the system, and the high degree of possible interactivity in the system. As a result, existing models/simulations must simplify the system by eliminating from consideration certain aspects, and therefore cannot be completely accurate in determining the effect of uncertainties on the performance of the NAS. To try to fill this gap in capability, a model of the NAS was developed, with the purpose of determining the effect of the various sources of uncertainty on the system performance of the NAS. This model is state-based, and models the system from a high level (Landry and Archer, 2013). The model is believed to be complete. Given this model, one can reason about the effects of uncertainty on the NAS, as well as the interactions of these sources of uncertainty (Landry and Archer, 2013). In addition, if one characterizes the effect of different operational concepts and technologies being considered under NextGen in terms of their effect on those sources of uncertainty, one can then reason about the relative benefits of those different concepts and technologies. In this work, we focus on the effect of concepts and technologies on the performance of the NAS. Following a brief discussion of the model, an enumeration of the technologies and concepts under consideration for NextGen will be presented, including their anticipated degree of effect on various sources of uncertainty. Lastly, the results of simulation exercises to test the model will be presented. II. Model of the NAS: Framework

The National Airspace System (NAS) is a large and complex system encompassing a wide-range of resources to enable safe and efficient air transportation. To effectively manage the NAS, it is very important to be able to reliably assess current and future demand for communications, navigation, and surveillance (CNS) resources as they are triggered by air traffic control (ATC) events. The authors present an integrated model that is capable of relating NAS ATC events to NAS CNS demand. Model validation with a subset of recorded ETMS CNS statistics will be presented. Combining such a model with NAS future demand predictors such as Terminal Area Forecasts (TAF) and Future Aviation Timetable Estimator (FATE), it becomes possible to estimate the increase in CNS demand and future requirements for CNS capacity, for resource planning purposes. The model presented in this paper includes tracking of radar-related messages and sector workload.

This paper proposes: 1) a traffic flow management policy, which allocates and organizes machine type communication (MTC) traffic flows network resources sharing within evolved packet system (EPS); 2) an access element as a wireless sensor network gateway for providing an overlaying access channel between the machine type devices and EPS; and 3) it addresses the effect and interaction in the heterogeneity of applications, services and terminal devices, and the related quality of service (QoS) issues among them. This paper overcomes the problems of network resource starvation by preventing deterioration of network performance. The scheme is validated through simulation, which indicates the proposed traffic flow management policy outperforms the current traffic management policy. Specifically, simulation results show that the proposed model achieves an enhancement in QoS performance for the MTC traffic flows, including a decrease of 99.45% in packet loss rate (PLR), a decrease of 99.89% in packet end to end (E2E) delay, a decrease of 99.21% in packet delay variation (PDV). Furthermore, it retains the perceived quality of experience of the real time application users within high satisfaction levels, such as the voice over long term evolution service possessing a mean opinion score (MOS) of 4.349 and enhancing the QoS of a video conference service within the standardized values of a 3GPP body, with a decrease of 85.28% in PLR, a decrease of 85% in packet E2E delay and a decrease of 88.5% in PDV.

As a result of the FAA Modernization and Reform Act of 2012, Unmanned Aerial Systems (UAS) are expected to be integrated into the National Airspace System (NAS) by 2015. Several human factors challenges need to be addressed before UAS can safely and routinely fly in the NAS with manned aircraft. Perhaps the most significant challenge is for the UAS to be non-disruptive to the air traffic management system. Another human factors challenge is how to provide UAS pilots with intuitive traffic information in order to support situation awareness (SA) of their airspace environment as well as a see-and-avoid capability comparable to manned aircraft so that a UAS pilot could safely maneuver the aircraft to maintain separation and collision avoidance if necessary. A simulation experiment was conducted to examine baseline compliance of UAS operations in the current airspace system. Researchers also examined the effects of introducing a Cockpit Situation Display (CSD) into a UAS Ground Control Station (GCS) on UAS pilot performance, workload and situation awareness while flying in a positively controlled sector. Pilots were tasked with conducting a highway patrol police mission with a Medium Altitude Long Endurance (MALE) UAS in L.A. Center airspace with two mission objectives: 1) to reroute the UAS when issued new instructions from their commander, and 2) to communicate with Air Traffic Control (ATC) to negotiate flight plan changes and respond to vectoring and altitude change instructions. Objective aircraft separation data, workload ratings, SA data, and subjective ratings regarding UAS operations in the NAS were collected. Results indicate that UAS pilots were able to comply appropriately with ATC instructions. In addition, the introduction of the CSD improved pilot SA and reduced workload associated with UAS and ATC interactions.

An Eulerian network model for air traffic flow in the National Airspace System is developed and used to design flow control schemes which could be used by Air Traffic Controllers to optimize traffic flow. The model relies on a modified version of the Lighthill-Whitham-Richards (LWR) partial differential equation (PDE), which contains a velocity control term inside the divergence operator. This PDE can be related to aircraft count, which is a key metric in air traffic control. An analytical solution to the LWR PDE is constructed for a benchmark problem, to assess the gridsize required to compute a numerical solution at a prescribed accuracy. The Jameson-Schmidt-Turkel (JST) scheme is selected among other numerical schemes to perform simulations, and evidence of numerical convergence is assessed against this analytical solution. Linear numerical schemes are discarded because of their poor performance. The model is validated against actual air traffic data (ETMS data), by showing that the Eulerian description enables good aircraft count predictions, provided a good choice of numerical parameters is made. This model is then embedded as the key constraint in an optimization problem, that of maximizing the throughput at a destination airport while maintaining aircraft density below a legal threshold in a set of sectors of the airspace. The optimization problem is solved by constructing the adjoint problem of the linearized network control problem, which provides an explicit formula for the gradient. Constraints are enforced using a logarithmic barrier. Simulations of actual air traffic data and control scenarios involving several airports between Chicago and the U.S. East Coast demonstrate the feasibility of the method.

This paper discusses the results of modeling and simulation performed by FAA’s Performance Analysis and Strategy office to forecast delays at thirty major US airports six months ahead of time. In order to have reliable delay forecasts, it is important that the simulation model validates well with historic actual data. As part of the validation efforts, we recognize that an aircraft’s movement in the US National Airspace System (NAS) involves its interaction with various resources along the flight path, such as airport runways, arrival and departure fixes, airspace sectors, and terminal gates. When multiple aircraft compete for access to these resources, their movements are impeded due to operational and safety considerations, resulting in flight delays. The objective of this study is to understand how various resources contribute to total flight delays, and how sensitive the delays are to changes in resource capacities. The results to date indicate that reduced airport runway capacities play a significant role in overall delays, but arrival and departure fixes and airspace sector constraints do not. Our use of historical airport AARs and ADRs for runway capacities, which may inherently account for fix constraints, yields delays that agree very well with actual values reported in ASPM. However, care has to be taken to use and interpret these rates when there are runway additions or closures. One area in which all NAS-wide models fail to replicate the behavior of the NAS is gate-out delay. Our simulation includes modeled distributions to capture gate-out delays, but they were found to capture only a fraction of actual gate-out delays. Terminal gate constraints were added to verify if they yielded large gate-in delays that would result in additional gate-out delays. No significant additional gate-out delays were obtained, although we gained better insight into how gate-in delays may occur. We are continuing to research potential NAS interactions that would account for the gap in predicting gate-out delays. Additional simulation model enhancements are planned to account for traffic flow initiatives such as GDP, as well as itinerary adjustments due to flight cancellation and substitutions.

One of the major issues in the proposed concepts-of-operations for NEXTGEN is the equitable allocation of overscheduled National Airspace System (NAS) resources. Although federal regulations, Congressional policies, and modernization plans call for equitable allocation of publicly held resources, the mechanisms for equitable allocation, and the trade-offs that must be made between stakeholders, exhibit a high degree of social, political, and economic complexity. Further, the allocations that are routinely made in Traffic Flow Management (TFM) and Air Traffic Control (ATC) are subject to widespread perceptions of systemic inequity and economic inefficiency. This paper describes the properties associated with the dynamic allocation of an overscheduled resource. The mathematical model demonstrates that (i) a natural asymmetry exists in the allocation, (ii) this asymmetry results in a low probability (<10%) of an equitable allocation of resources amongst groups of flights (e.g. airlines), and (iii) increased competition reduces the likelihood of equitable allocations. These theoretical results establish the ldquofeasible spacerdquo in which claims of equity can be made, suggests refinements of the existing allocation schemes, and establishes the framework for a regulatory role in proposed market-based mechanisms.