Conflict Resolution Maneuvers Research Articles

Deep reinforcement learning based path stretch vector resolution in dense traffic with uncertainties

With the continuous growth in the air transportation demand, air traffic controllers will have to handle increased traffic and consequently, more potential conflicts. This gives rise to the need for conflict resolution advisory tools that can perform well in high-density traffic scenarios given a noisy environment. Unlike model-based approaches, learning-based approaches can take advantage of historical traffic data and flexibly encapsulate environmental uncertainty. In this study, we propose a reinforcement learning approach that is capable of resolving conflicts, in the presence of traffic and inherent uncertainties in conflict resolution maneuvers, without the need for prior knowledge about a set of rules mapping from conflict scenarios to expected actions. The conflict resolution task is formulated as a decision-making problem in a large and complex action space. The research also includes the development of a learning environment, scenario state representation, reward function, and a reinforcement learning algorithm inspired from Q-learning and Deep Deterministic Policy Gradient algorithms. The proposed algorithm, with two stages decision-making process, is used to train an agent that can serves as an advisory tool for air traffic controllers in resolving air traffic conflicts where it can learn from historical data by evolving overtime. Our findings show that the proposed model gives the agent the capability to suggest high quality conflict resolutions under different environmental conditions. It outperforms two baseline algorithms. The trained model has high performance under low uncertainty level (success rate ≥95% ) and medium uncertainty level (success rate ≥87%) with high traffic density. The detailed analysis of different impact factors such as environment’s uncertainty and traffic density on learning performance are investigated and discussed. The environment’s uncertainty is the most important factor which affects the performance. Moreover, the combination of high-density traffic and high uncertainty will be the challenge for any learning models.

Why do Controllers Choose the Conflict Resolution Maneuvers that They Do?

ABSTRACT Objective: This study aimed to reveal the causal reasons for maneuver preferences on the basis of the objective measures. Background: Although the effects of conflict geometry have been well documented in previous literature, empirical evidence on the influence of conflict geometry on the operator’s workload, maneuver choice, and success is still lacking. Method: Thirteen undergraduate students with air traffic control (ATC) knowledge were requested to complete ATC tasks and resolve six different conflicts that were generated by manipulating two geometric features, namely, lateral (i.e., crossing, converging, and overtaking) and vertical (i.e., level and nonlevel) conflicts. Results: Crossing (p < .01) and converging (p = .05) conflicts resulted in higher workload than overtaking conflicts because of the higher geometric dimensions involved. Workload (p < .01) and performance (p < .03) were worse during nonlevel conflicts than during level conflicts. Notably, vertical maneuver led to a higher workload than other maneuver choices despite the higher preference for this maneuver. These findings were associated with visualization load. Conclusion: We failed to confirm that the resolution maneuver was chosen because of its good performance and low workload. Instead, predetermined rules (i.e., altitude, speed, and heading) were used, regardless of the workload, and the safety of the chosen maneuver for a particular conflict geometry was verified.

RPAS Automatic ADS-B Based Separation Assurance and Collision Avoidance System Real-Time Simulation Results

Remotely piloted aircraft systems (RPAS) are increasingly becoming relevant actors that are flying through the airspace and will gain much more importance in the future. In order to allow for their safe integration with manned conventional traffic in non-segregated airspaces, in accordance with the overall air traffic management (ATM) paradigm, specific enabling technologies are needed. As is well known, the detect and avoid (DAA) technology is fundamental among the enabling technologies identified as crucial for RPAS integration into the overall ATM system. In the meantime, to support extended surveillance, the universal introduction of cooperative automatic dependent surveillance-broadcast (ADS-B) on-board aircraft is being increasingly implemented because it has the potential to allow for the coverage of the entire airspaces in remote areas not usually covered by conventional radar surveillance. In this paper, experimental results that were obtained through the real-time validation, with hardware and human in the loop (RTS-HIL) simulations, of an automatic ADS-B based separation assurance and collision avoidance system aimed to support RPAS automatic operations (as well as remote pilot decision making) are presented and discussed. In the paper, after an introductory outline of the concept of operations (ConOps) of the system and its architectural organization, in addition to basic information about the main system functionalities, a description of the tests that were carried out is reported, and the obtained results are described and discussed in order to emphasize the performance and limitations of the proposed system. In particular, the obtained quantitative performances are reported and commented on, and the feedback presented by pilots in order to improve the system, e.g., in terms of preferred typology of conflict resolution maneuver elaborated by the system, is described.

Nonlinear Conflict Resolution and Flow Management Using Particle Swarm Optimization

A new optimization problem to resolve conflicts by allowing aircraft to change both their heading angle and speed is considered. The performance index in the optimization problem is formulated to reduce the variation of the aircraft arrival time that is caused by conflict resolution maneuvers. To achieve both conflict resolution and flow management, metering constraints are introduced together with separation constraints. Even though the performance index and constraints are nonlinear, the optimal solution can be easily obtained by utilizing particle swarm optimization. Numerical simulations are performed to evaluate the performance of the proposed algorithm, and the numerical experimental results showed a significant reduction in the variation of the aircraft arrival time as well as the magnitude of heading angle and speed changes.

Path Planning for Conflict Resolution in Free Flight System: Optimization Based on Linear Programming

Air traffic congestion is one of the major challenges being faced by current air traffic management system all over the world. Congestion is mainly due to the incapability of radar based air traffic controllers to meet today’s increased air transportation demands. Free flight concept which provides each aircraft the freedom to self optimize their routes has been recently proposed as solution to the problem of air traffic congestion. Avionics community is now working towards the realization of this concept. Conflict detection between aircraft and its subsequent resolution are the challenging tasks in the implementation of free flight concept. Once a conflict is detected, it can be resolved by altering the aircraft nominal route by changing aircraft’s velocity or/and heading angle and when the conflict is avoided, the aircraft may return to its original optimal path. Also it is desirable to do all these path changes in minimum time so as to minimize the delay and fuel consumption. An efficient path planning algorithm is thus required to optimize the changes in velocity and heading angle that must be followed by the aircraft during conflict resolution maneuvers. In this paper, a path planning algorithm based on linear programming is adopted to ensure conflict free path in minimum time. With this algorithm the autopilot can resolve the conflict effectively. The effectiveness of the algorithm is valiadted through simulation studies.

Hybrid systems modeling and automated air traffic control for three-dimensional separation assurance

The increasing volume of traffic in the air transportation is leading to excessive workload on air traffic controllers. Developing automated air traffic management (ATM) tools is a critical technology in reducing the workload of air traffic controllers and hence increasing the airspace capacity. The existing approaches to automated ATM either use overly-simplified air traffic and aircraft dynamics models to reduce computational complexity or end up being computationally intractable for large-scale ATM scenarios. This paper presents a new hybrid system description of modeling the decision process of the air traffic controllers in en-route and approach operations. The model is based on the domain expertise provided by the State Airport Authority and Air Navigation Service Provider (ANSP) of Turkey. The emulation of air traffic controller decision process in the hybrid model provides realistic conflict resolution maneuvers and separation assurance in 3D, while being computationally tractable. The algorithm has polynomial iteration complexity in the number of waypoints of the aircraft, which makes it scalable to large-scale ATM scenarios with more than 100 aircraft. The algorithm is validated on the real air traffic data over the Istanbul region extracted from the ALLFT+ dataset provided by EUROCONTROL, which includes over 7000 flights in a 96-hour period. The developed algorithm is also integrated into a Boeing 737-800 flight deck simulator with a custom radar display to demonstrate the applicability to existing avionics systems.

Effects of UAS Performance Characteristics, Altitude, and Mitigation Concepts on Aircraft Encounters and Delays

Introduction of Unmanned Aircraft Systems (UAS) into positively controlled airspace may cause greater per-aircraft impact on the National Airspace System than existing traffic because UAS aerodynamic performance and mission types are often different from aircraft that typically fly in positively controller airspace (i.e. commercial passenger traffic). This paper examines the impact new UAS operations will have on existing aircraft and measures that impact by the number of predicted conflicts and associated conflict resolution delays that occurred in fast-time simulation. The two conflict metrics are quantified as a function of the UAS altitude and cruise speed. Two mitigation approaches are also investigated: increasing horizontal separation requirements for UAS and “burdening” UAS with the responsibility to execute all resolution maneuvers when possible. Results indicate that en route conflict maneuver delay for existing traffic because of new UAS operations can be nearly eliminated by burdening UAS with the responsibility to execute conflict resolution maneuvers and maintaining the current en route horizontal separation requirement of five nautical miles from other aircraft.

Conflict Resolution Maneuvers in Air Traffic Control: Investigation of Operational Data

Little is known about air traffic controllers' preferred maneuvers to resolve conflicts between pairs of aircraft, despite the topic's importance of to the planned NextGen system. We analyzed 256 cases of conflict resolution maneuvering from radar track data from 5 U.S. air traffic control centers. Our results show the influences of conflict geometries on controllers' maneuver choices.

Near Real-Time Fuel-Optimal En Route Conflict Resolution

In this paper, we consider the air-traffic conflict-resolution problem and develop an optimization model to identify the required heading and speed changes of aircraft to avoid conflict such that fuel costs are minimized. Nonconvex fuel functions in the optimization problem are modeled through tight linear approximations, which enable the formulation of the problem as a mixed-integer linear program. The significance of the developed model is that fuel-optimal conflict-resolution maneuvers can be identified in near real time, even for conflicts involving a large number of aircraft. Computational tests based on realistic air-traffic scenarios demonstrate that conflicts involving up to 15 aircraft can be solved in less than 10 s with an optimality gap of around 0.02%.

Are There Standard Conflict Resolution Maneuvers in Air Traffic Control?

We analyzed samples of aircraft track data involving conflict alerts and subsequent resolution maneuvers from five U.S. air route traffic control centers. Vertical conflict resolution maneuvers were used in the majority of the cases examined. Within the vertical dimension, reductions of current vertical change (climb or descent) were collectively the most frequent resolution maneuver type, but descents were about three times as frequent as climbs. Three contingency analyses were performed to test for independence between conflict geometries and subsequent resolution maneuvers. The results show that conflict resolution maneuvers do not seem to be independent from conflict geometries. Additional analyses on conflict angles and vertical speed differences between aircraft as continuous variables revealed disproportionately fewer turns on crossing courses than on opposite or same courses. These results provide critical baseline information for understanding just how large departures from the current ingrained practices the advent of the NextGen air traffic management infrastructure may bring about, particularly in extremely safety-critical tasks such as airborne conflict resolutions. Pilots' conflict resolution maneuver preferences have received some attention, but corresponding research on air traffic controllers' practices is almost nonexistent. Moreover, our results emerged from objective data through objective analyses. To our knowledge, this is the first time analyses like these have been applied to data like these.

Display Dimensionality and Conflict Geometry Effects on Maneuver Preferences for Resolving In-Flight Conflicts

Two experiments explored the effects of display dimensionality, conflict geometry, and time pressure on pilot maneuvering preferences for resolving en route conflicts. With the presence of a cockpit display of traffic information (CDTI) that provides graphical airspace information, pilots can use a variety of conflict resolution maneuvers in response to how they perceive the conflict. Inconsistent preference findings from previous research on conflict resolution using CDTIs may be attributable to inherent ambiguities in 3-D perspective displays and/or a limited range of conflict geometries. Pilots resolved predicted conflicts using CDTIs with three levels of display dimensionality; the first had two 2-D orthogonal views, the second depicted the airspace in two alternating 3-D perspective views, and the third had a pilot-controlled swiveling viewpoint. Pilots demonstrated the same preferences that have been observed in previous research for vertical over lateral maneuvers in low workload and climbs over descents for level-flight conflicts. With increasing workload the two 3-D perspective displays, but not the 2-D displays, resulted in an increased preference for lateral over vertical maneuvers. Increased time pressure resulted in increased vertical maneuvers, an effect again limited to the two 3-D perspective displays. Resolution preferences were more affected by workload and time pressure when the 3-D perspective displays were used, as compared with the 2-D displays, although overall preferences were milder than in previous studies. Investigating maneuver preferences using the strategic flight planning paradigm employed in this study may be the key to better ensure pilot acceptance of computer-generated resolution maneuvers.

Comparison of Pilots' and Controllers' Conflict Resolution Maneuver Preferences

The role of the air traffic controller in future air traffic management systems is that of a passive monitor, who intervenes in potential conflict situations only if pilots, aided by onboard traffic displays, fail to implement safe and timely conflict resolutions. Yet, such scenarios are entirely plausible and even likely, particularly in busy airspaces. However, it is also likely that an a controller will seek to implement a resolution that is different from what the pilots had planned, resulting in confusion, delays, and safety risks. This paper examines pilots' and controllers' maneuver preferences and potential sources for dissonance in conflict resolution situations. Data on pilots' maneuver tendencies were gathered from an extensive review of past literature; controllers' preferences were obtained from an experiment that systematically manipulated simulated ATC scenario difficulty and conflict geometries. Results indicate largely similar maneuver choices between controllers and pilots, but suggest a need for further research.

Display Dimensionality and Conflict Geometry Effects on Maneuver Preferences for Resolving in-Flight Conflicts

With the presence of a cockpit display of traffic information (CDTI) that provides graphical airspace information, pilots can use a variety of conflict resolution maneuvers in response to how they perceive the conflict. Inconsistent preference findings from previous research on conflict resolution using CDTIs may be due to inherent ambiguities in 3-D displays and/or a limited range of conflict geometries. This paper describes a study that investigates conflict resolution maneuver preferences using three displays with different frames of reference and a wide range of conflict geometries. Results indicate that 3-D displays with interactive viewpoints reduced spatial ambiguities. The interactive 3-D displays produced a preference for vertical maneuvers over lateral similar to a 2-D coplanar display; however this preference was reversed under increased workload conditions for both 3-D displays. Pilots in all three display conditions showed a preference to maneuver vertically away from intruders, though this was eliminated (or reversed) as workload increased. The 2-D coplanar display induced a preference to laterally turn away from approaching intruders, which overwhelmed a trained “turn right” preference.

Safety verification of conflict resolution manoeuvres

We address the problem of generating provably-safe conflict resolution maneuvers for aircraft in uncertain environments. We assume that a maneuver is composed of a sequence of flight modes, which are segments of constant heading, of constant bank angle, or of constant airspeed. Each of these flight modes has associated to it the kinematics of the aircraft, and hence the maneuver is a hybrid system. While the flight modes are defined ahead of time, their sequencing and parameter values do not necessarily have to be. We present an algorithm for generating provably safe maneuvers, which is based on a general procedure for designing controllers for hybrid systems. The result is a maneuver, proven to be safe within the limits of the models used, which is a familiar sequence of commands easily executable by the flight management systems. The maneuvers may be viewed as protocols, or rules of the road, and are well-defined for each conflict scenario. We present results for two example maneuvers.

Methodology for the Performance Evaluation of a Conflict Probe

A conflict probe is an air traffic management tool that can predict conflicts well in advance, using information on aircraft position, speed and plans along with forecasts of wind and temperature profiles. Strategic (early) resolution of conflicts has the potential to reduce the cost associated with conflict resolution maneuvers. A conflict probe would be especially useful in a free flight environment, which is expected to have a less structured traffic flow compared to the current operating environment. This paper presents a comprehensive methodology to quantitativel y evaluate the performance of a conflict probe, using real traffic data and expanded separation criteria. The methodology is generic in nature, and can be applied to any conflict probe. It can therefore provide a framework for a comparative study of conflict probes. Several metrics of conflict probe performance have been developed and evaluated. The missed alert rate and false alert rate are primary metrics that quantify the reliability of a conflict probe; the mean conflict warning time and errors in key conflict prediction parameters such as minimum horizontal and vertical separations are important secondary metrics that quantify the accuracy of a conflict probe. Work is underway to validate the evaluation methodology developed in this work by applying it to the Center-TRACON Automation System's Conflict Probe Tool, using real traffic data from the Denver Air Route Traffic Control Center; some preliminary results are presented as an example.

Conflict Resolution Maneuvers during near Miss Encounters with Cockpit Traffic Displays

This experiment investigated the benefits and liabilities associated with pilots use of a cockpit traffic display to assess the threat posed by other air traffic, and if necessary, to make small maneuvers to avoid situations which would otherwise result in collision avoidance advisories. The main experimental variables were the amount of and quality of the information that was displayed on the intruder aircraft's estimated future position. The crew's task was to fly a simulation of a wide body aircraft along a straight course at a constant altitude. Intruder aircraft appeared on a variety of converging trajectories. When the intruder was 60 seconds from the point of closest approach, the pilots were to maneuver so as to maintain a horizontal separation of at least 1.5 nm or a vertical separation of 500 ft. With this separation a CLIMB or DIVE collision avoidance advisory would not be triggered. As a secondary objective, the pilots were instructed to attempt to stay within 500 ft and 1.5 nm of their assigned route. The results show that the pilots could usually make maneuvers that would provide the specified separation but that in doing so they often made course deviations greater than 1.5 nm or 500 ft.