Gonzalo de Miguel

Trajectory classification based on machine-learning techniques over tracking data

This work addresses the application of a machine-learning approach to classify ATC trajectory segments from recorded opportunity traffic. It is based on the mode probabilities estimated by an IMM tracking filter operating forward and backward over available data. A learning algorithm creates a rule base for classification from these data, once they have been properly prepared. Performance of this data-driven classification system is compared with a more conventional approach based on transition detection on simulated and real data of representative situations. The offline processing of real data allows an accurate classification of manoeuvring segments, with the possibility of synthesizing ground truth lines for performance evaluation

ATC trajectory reconstruction for automated evaluation of sensor and tracker performance

Currently most air traffic controller decisions are based on the information provided by the ground support tools provided by automation systems, based on a network of surveillance sensors and the associated tracker. To guarantee surveillance integrity, it is clear that performance assessments of the different elements of the surveillance system are necessary. Due to the evolution suffered by the surveillance processing chain in the recent past, its complexity has been increased by the integration of new sensor types (e.g., automatic dependent surveillance-broadcast [ADS-B] [1], Mode S radars [2], and wide area multilateration [WAM] [3]), data link applications, and networking technologies. With new sensors, there is a need for system-level performance evaluations as well as methods for establishing assessment at each component of the tracking evaluation.

Model-based trajectory reconstruction with IMM smoothing and segmentation

Abstract This paper presents a new approach for off-line trajectory reconstruction in air traffic control domain. The proposed algorithm, called model-based reconstruction, performs an accurate IMM smoothing process whose parameters are modified along time according to the flight modes segmented from trajectory measurements. Its competitive performance is demonstrated through comparison with previous reconstruction methods used in ATC and with classical IMM smoothing, using simulated data.

Adaptive data fusion for air traffic control surveillance

This paper describes a method to enhance current surveillance systems used in air traffic control. Those systems are currently based on statistical data fusion, relying on a set of statistical models and assumptions. The proposed method allows for the on-line calibration of those models and enhanced detection of non-ideal situations, increasing surveillance products integrity. It is based on the definition of a set of observables from the fusion process and a rule based expert system with the objective to change processing order, algorithms or even remove some sensor data from the processing chain.

Traffic analysis and synthetic scenario generation for ATM operational concepts evaluation

This paper describes a pair of systems which can be used to obtain realistic traffic samples in a Sector/TMA from a given real traffic database. Those are a Traffic Analyzer and a Traffic Pattern Generator. These two systems allow the ATM engineer to both gain insight on the traffic structure of the area under analysis and to obtain statistically significant samples for the evaluation of operational concepts and procedure changes, perform analysis of ATM performance under traffic changes, …

A new approach to map-assisted Bayesian tracking filtering

Abstract This paper presents a new non-linear filter designed to track targets following a road network, taking advantage of the road map information. The algorithm is based on a Bayesian Multiple Hypotheses modelling of movement process, postulating and evaluating different hypotheses on the segments being followed by the target after road junctions. Then, the along-road tracking is carried out, for each hypothesis, by a longitudinal IMM filter capable of tracking target movements along straight roads, circular segments, and generic curvilinear segments defined through Bezier curves. The algorithm also includes a lateral drift estimator, which tracks the lateral motion of the target with respect to road axis, to be able to estimate target piloting error and especially to track targets in wide roads. The paper completely describes the filter and associated measurement preprocessing procedures, and also includes a comparative evaluation of the proposed filter with other filtering methods in the literature.

Complete Systematic Error Model of SSR for Sensor Registration in ATC Surveillance Networks

In this paper, a complete and rigorous mathematical model for secondary surveillance radar systematic errors (biases) is developed. The model takes into account the physical effects systematically affecting the measurement processes. The azimuth biases are calculated from the physical error of the antenna calibration and the errors of the angle determination dispositive. Distance bias is calculated from the delay of the signal produced by the refractivity index of the atmosphere, and from clock errors, while the altitude bias is calculated taking into account the atmosphere conditions (pressure and temperature). It will be shown, using simulated and real data, that adapting a classical bias estimation process to use the complete parametrized model results in improved accuracy in the bias estimation.

Air traffic generation for new terminal area air traffic management concepts design and evaluation

This paper describes two systems that can be used to obtain realistic random traffic samples in a terminal area: a real traffic analyser and a synthetic traffic generator. These two systems allow the air traffic management (ATM) engineer to gain insight on the traffic structure of the area under analysis, and allow obtaining realistic traffic samples enabling the evaluation of new operational concepts, the validation or system performance measurement after procedure changes, the analysis of ATM performance under forecasted future traffic changes, etc. Together with the design of the system, the work provides insight of user interfaces and describes the potential uses of such tools in an integrated ATM system.

Performance requirements specification design for Airport Surface Surveillance

This paper describes a method to define the performance requirements for Airport Surface Surveillance. The key idea is making the performance specification dependent on the underlying sensor deployment and geometric definition of the scenario, which enables its extension to any operational deployment. The described methodology was used in SESAR technical specifications to provide a shared view of the desired performance for users and industry.

Air-to-air surveillance for future ATM systems

This paper describes a prototype implementation of ADS-B in an air-to-air surveillance system. This air surveillance system was defined for an experimental project on future civilian air traffic management, which imposes new requirements over both air and ground surveillance systems. In this paper the relation between the different airborne and ground systems related to air surveillance is detailed, and the air surveillance and its algorithms are described. It is related with current ADS-B and TIS-B proposals, and defined as an extension of those systems. Finally, performance of the air surveillance is detailed, based on simulated scenarios, and some conclusions for future enhancements of ADS-B function are derived.