DART (Data-driven AiRcraft Trajectory prediction research) addresses the topic “ER-02-2015 – Data Science in ATM” exploring the applicability of data science and complexity science techniques to the ATM domain. DART delivers an understanding on the suitability of applying big data and agent –based modelling techniques for predicting aircraft trajectories based on data-driven models and accounting for ATM network complexity effects, considering multiple correlated trajectories.
DART has been motivated by the fact that the current Air Traffic Management (ATM) system worldwide has reached its limits in terms of predictability, efficiency and cost effectiveness. Nowadays, the ATM is based on an airspace management paradigm that leads to demand imbalances that cannot be dynamically adjusted. This entails higher air traffic controllers’ workload, which, as a final result, determines the maximum system capacity.
With the aim of overcoming the ATM system deficiencies, different initiatives, dominated by SESAR in Europe and Next Gen in the US, have promoted the transformation of the current environment towards a new trajectory-based ATM paradigm. This paradigm-shift changes the old-fashioned airspace management to the advanced concept of Trajectory Based Operations (TBO).
[Intentionally emphasized] The proposed transformation requires high-fidelity aircraft trajectory prediction capabilities, supporting the trajectory life-cycle at all stages efficiently. Making accurate predictions about individual trajectories in an early phase of operations, should allow predicting ATM network status, evolution of demand for resources, and thus hotspots, accounting for complex phenomena of the ATM system, when predictions are combined. Thus, predictions should allow assessing the overall system status and the impact of traffic on individual trajectories, w.r.t. operational constraints. Consequently, advances towards this direction should support effective decision making and optimization of resource exploitation during operations time. Motivated by these objectives, and given the existing and growing wealth of ATM data, as well as advances in machine learning, DART focuses on data-driven approaches to increasing predictability, and agent-based approaches for accounting for complex phenomena in the overall ATM system due to traffic and congestion of resources: These are major areas in which DART developments contribute.
Research topic
- Air Traffic Management
- Demand-Capacity Imbalance Resolution
- Multiagent Reinforcement Learning
- Trajectory Prediction
- Congestions Resolution
Duration
2016 – 2018
Objectives
- Definition of requirements for the input datasets needed. The requirements will consider the trajectory prediction accuracy expected.
- Study of the application of big-data techniques to trajectory related data gathering, filtering, storing, prioritization, indexing or segmentation to support the generation of reliable and homogenous input datasets.
- Study of different data-driven learning techniques to describe how a reliable trajectory prediction model will leverage them.
- Formal description of the complexity network to support correlated multiple trajectory predictions .
- Study of the application of agent-based models to the prediction of multiple correlated trajectory predictions considering complexity network.
- Description of visualization techniques to enhance trajectory data management capabilities.
- Exploration of advanced visualization processes for data-driven model algorithms formulation, tuning and validation, in the context of 4D trajectories.
Leave a Reply
You must be logged in to post a comment.