In short




2016-06-20 > 2018-06-19


598 523


598 523



Trajectory prediction – letting the machine to the math

The complex ATM system worldwide has reached its limits in terms of predictability, efficiency and cost effectiveness. The DART explored the potential of data-driven techniques for trajectory prediction, and agent-based modelling approaches for assessing the impact of traffic to individual trajectories, thus accounting for ATM complex phenomena. Improvements with consequent benefits in these emerging areas of research can support the trajectory-based operations (TBO) paradigm.

The project focused on providing answers to the following major questions:

• What are the supporting data required for accurate trajectory predictions?

• What is the potential of machine learning algorithms to support high-fidelity aircraft trajectory prediction?

• How the complex nature of the ATM system impacts trajectory predictions?

• How can this insight be used to optimise the ATM system?

DART explored the potential of machine learning methods using historical data to increase the predictability for individual trajectories, and multi-agent collaborative reinforcement learning methods to resolve demand-capacity balancing (DCB) problems, supporting the incorporation of stakeholders’ preferences into the planning process.

Results suggest that data-driven methods, compared to model-based approaches, can enhance trajectory prediction capabilities by exploiting patterns derived from historical data. In addition to that, agent-based methods can regulate flights effectively, reducing imposed delays, while resolving DCB problems.

The DART developments pave the way towards advanced collaborative decision-making processes that support multi-objective optimisation taking the requirements of the different stakeholders in the ATM system into account at the planning phase.

DART delivered machine learning techniques to improve the accuracy of trajectory predictions, accounting for ATM network complexity effects.

DART shapes the future towards collaborative decision-making processes among the stakeholders in the ATM system.


  • Increased predictability
  •  Reduced delays thanks to early resolution of DCB issues
  • Advancing collaborative-decision making and planning tools

Project Members:  University of Piraeus (Coordinator), Fraunhofer Gesellschaft, Boeing Research and Technology Europe  (BRTE), CRIDA

This project has received funding from the SESAR Joint Undertaking under the European Union's Horizon 2020 research and innovation programme under grant agreement No 699299

European Union