Jian Wei

Multi-Sensor Fusion and Fault Detection using Hybrid Estimation for Air Traffic Surveillance

Data fusion for multiple surveillance sensors in air traffic control (ATC) is studied. The goal is to build up software redundancy for better target tracking accuracy and robustness against sensor faults. A set of hybrid estimation algorithms for different sensors is designed to run in parallel for tracking aircraft with changing flight modes. The proposed sensor fusion algorithm combines the estimates from each hybrid estimation algorithm and identifies potential sensor faults.

A strategic conflict avoidance approach based on cooperative coevolutionary with the dynamic grouping strategy

Conflict avoidance plays a crucial role in guaranteeing the safety and efficiency of the air traffic management system. Recently, the strategic conflict avoidance (SCA) problem has attracted more and more attention. Taking into consideration the large-scale flight planning in a global view, SCA can be formulated as a large-scale combinatorial optimisation problem with complex constraints and tight couplings between variables, which is difficult to solve. In this paper, an SCA approach based on the cooperative coevolution algorithm combined with a new decomposition strategy is proposed to prevent the premature convergence and improve the search capability. The flights are divided into several groups using the new grouping strategy, referred to as the dynamic grouping strategy, which takes full advantage of the prior knowledge of the problem to better deal with the tight couplings among flights through maximising the chance of putting flights with conflicts in the same group, compared with existing grouping strategies. Then, a tuned genetic algorithm (GA) is applied to different groups simultaneously to resolve conflicts. Finally, the high-quality solutions are obtained through cooperation between different groups based on cooperative coevolution. Simulation results using real flight data from the China air route network and daily flight plans demonstrate that the proposed algorithm can reduce the number of conflicts and the average delay effectively, outperforming existing approaches including GAs, the memetic algorithm, and the cooperative coevolution algorithms with different well-known grouping strategies.

Multilevel Graph Partitioning Algorithm for Dynamic Airspace Configuration

In this paper, a multilevel graph partitioning algorithm is developed for Dynamic Airspace Configuration which partitions the graph model of airspace with given user defined constraints and hence provides the user more flexibility and control over various partitions. We formulate the airspace configuration problem as a graph partitioning problem to balance the sub-graph (sector) workload while satisfying the capacity constraint by using a graph model which accurately represents the air route structure and air traffic in the National Airspace system (NAS). Currently the spectral clustering method is widely in use for such type of partitioning problems. However it does not provide the user the flexibility to take into consideration the structure of the graph (e.g. if a group of vertices needs to be in the same group, if a particular edge or link should not be allowed to be cut etc.). In terms of air traffic management, vertices represent airports and waypoints. Some of the major (busy) airports need to be given more importance and hence treated separately. Thus our algorithm takes into account the air route structure while finding a balance between sector workloads. Since graph partitioning algorithms usually face problems such as disconnected sub-graphs and unbalanced partitions, an algorithm is proposed to refine these partitions. The performance of the proposed algorithm is validated with Enhanced Traffic Management System (ETMS) air traffic data.

An Integer Programming based Sector Design Algorithm for Terminal Dynamic Airspace Configuration

The future air travel demand will keep increasing at a steady rate, and the future flights and trajectory-based operations will become more flexible. Due to lack of flexibility, the current National Airspace System (NAS) does not have the ability to efficiently cope with the increasing air travel demand or to implement flexible use of future airspace. A lot of work has been done in the framework of Dynamic Airspace Configuration (DAC), aiming at dynamically configuring airspace to better accommodate the fluctuating air traffic demand and changing traffic pattern, but most past research has been focused on en route airspace. In this paper, a sector design algorithm is proposed for DAC in the terminal airspace, which combines a constrainted k-means clustering algorithm, integer programming techniques, and an alpha shapes based sectorization algorithm to improve the efficiency of airspace utilization and reduce the traffic complexity. Historical traffic data from major U.S. airports such as Hartsfield-Jackson Atlanta International Airport (ATL) and Dallas/Fort Worth International Airport (DFW) and simulated traffic data with future route designs and traffic patterns are used to validate the proposed algorithm and evaluate its performance.

Preliminary assessment of operational benefits for a graph-based Dynamic Airspace Configuration algorithm

Previously, a graph-based algorithm for essence of Dynamic Airspace Configuration (DAC) is developed, in alliance with the essential ideas of the DAC concepts: innovative air space structure and time-varying adaptation to those structures. This paper presents preliminary assessment of operational benefits for the DAC algorithm using ETMS data of the Cleveland center (ZOB) for a good weather day. The airspace of ZOB is vertically divided into two fields: high sectors (24000ft and above) and low sectors (10000ft to 23900ft). Moreover, sector boundaries are generated algorithmically in intervals of 60 minutes and 30 minutes for the purpose of inherent exploration of the air traffic structure. Mainly, three metrics are considered: number of sectors, aircraft counts in a sector, and dwell time. Preliminary results indicate benefit for staffing requirement and performance robustness to time-varying data of our algorithm. Inversely, through the assessment results our algorithm exhibits intensive support for DAC concepts.

Mathematical Programming Based Algorithm for Dynamic Terminal Airspace Configuration

distribution, alleviate the demand-capacity imbalances, and thereby increase the throughput of the entire national airspace system. However, past research has mainly focused on en route airspace. In this paper, we propose a mathematical programming based algorithm for dynamic terminal airspace conguration. With the understanding of the current terminal airspace organization and terminal operations, we handle the altitude changes by vertical stratication, formulate the airspace sectorization problem as an optimization problem, include separation regulations, ight path requirements and geometric restrictions as constraints, and construct a mathematical program. By solving for its optimal solution, we generate terminal airspace sectors which can account for the major trac ows and accommodate the trac variations. A preliminary benet analysis is conducted to show the promising benets of the proposed algorithm.

Diagnostic Throughput Factor Analysis Tool for En-Route Airspace

Today’s National Airspace System (NAS) is facing a challenge of dealing with an increasingly larger number of flight operations. To address this, the Next Generation Air Transportation System (NextGen) is introduced by the Federal Aviation Administration (FAA) to improve the efficiency and safety of the NAS. Currently, the performance of the NAS is mainly measured using the delay-based metrics that cannot capture the positive aspects of the performance and level of utilization of the NAS. To address this issue, a diagnostic throughput factor analysis tool is proposed to analyze and quantify the factors that have greater responsibility for poor/better regions/times of performance of the en-route airspace using the concept of throughput. Through a function-level comparison with other applications such as ground transportation, manufacturing, and wireless communication, major factors affecting the NAS’s throughput performance are identified as Monitor Alert Parameter violation, aircraft conflict, metering, c...

Estimated Time of Arrival Prediction based on State-Dependent Transition Hybrid Estimation Algorithm

The essential functions of air traffic management involve predicting and controlling the spatial and time relationships of aircraft at critical waypoints such as metering fixes and feeder gates. In the descent stage, for aircraft entering the terminal airspace, the Descent Advisor, a decision support tool in the current practice, assists the controllers to generate advisories for merging, sequencing, and separating while providing assistance for the prediction and resolution of conflicts. With the implementation of the Next Generation Air Transportation System, more accurate aircraft position information will be available, which lays the foundation for more accurate and reliable positioning, navigation and tracking, and therefore provides the basis for 4-dimensional trajectory based operations. In this paper, an aircraft’s trajectory estimation and estimated time of arrival (ETA) prediction algorithm is proposed. The proposed algorithm uses a stochastic nonlinear hybrid system to accurately model the aircraft descend procedure and the State-Dependent-Transition Hybrid Estimation method for ETA prediction. The proposed algorithm is demonstrated with a continuous descent approach example.

Benefit Analysis of a Sector Design Algorithm for Terminal Dynamic Airspace Configuration

As demand for air travel increases at a staggering rate, the National Airspace System (NAS) is becoming unable to effectively handle the increased stress on the system. To solve these inefficiencies, the Next Generation Air Transportation System (NextGen) has been proposed to improve the adaptability of the airspace. Because of the importance of the terminal airspace to the overall efficiency of the system, dynamically adapting the terminal airspace is crucial for adapting to future traffic patterns. An algorithm has been developed to use air routes and traffic data to dynamically generate sectors for various configurations and airports. This paper will evaluate the sectors created by our sector design algorithm for terminal dynamic airspace configuration (TDAC) using multiple methods. The first method will compare the performance of sectors generated with real traffic data and that of the current sectors. Using metrics inspired by dynamic density components, the traffic complexity can be estimated from current and algorithm generated sectors to quantify the algorithm’s success. The sector design algorithm will then be applied to future, simulated scenarios to determine if it will generate consistent results between real and simulated data. Finally, the effect of dynamic airspace configurations will be evaluated by analyzing the benefit of dynamical resectorization for various time intervals.

Design and Evaluation of a Dynamic Sectorization Algorithm for Terminal Airspace

Given its static and rigid structure, the current National Airspace System lacks the ability to cope efficiently with the increasingly severe demand–capacity imbalances expected to develop over the coming years. To better accommodate the flexibility desired for future flight operations and to alleviate the demand–capacity imbalances, research initiatives have been conducted under the dynamic airspace configuration concept. Although most past dynamic airspace configuration researchers have focused on en route airspace, this paper investigates terminal airspace operations. A dynamic sectorization algorithm for terminal airspace is developed, which combines a k-means clustering-based vertical sectorization algorithm, an integer-programming-based horizontal sectorization algorithm, and an α-shapes-based airspace sectorization algorithm. This dynamic sectorization algorithm is validated with real traffic data from several major international airports in the United States and simulated traffic data with project...