This deliverable presents support algorithms for automated tug assignment and multiagent path planning. These algorithms were designed on the basis of the algorithms for tug assignment and path planning developed in SESAR AEON project. The former algorithms were extended to take into account diverse and realistic spatiotemporal constraints related to airport surface movement operations (such as airport’s traffic rules) and the ASTAIR concept in particular. The developed algorithms dynamically adapt to changes in the environment (such as changes of runway mode of operation) and new constraints provided by human operators by recalculating their solutions. To address scalability issues of the previously developed tug assignment algorithms and to make them suitable for real-time use, an efficient, meta-heuristics-based approach was developed using Ant Colony Optimisation. The multiagent path planning algorithm developed in AEON was further improved by taking into account more realistic details (such as shapes and improved kinematic models of the vehicles) and enhancing its computational efficiency. During the workshops and interviews conducted in ASTAIR (WP1), preferred interactions of human operators with automated systems and algorithms were identified and described in deliverable D1.2 in the context of eight use cases. In this deliverable, we describe how some of these interactions were modelled and implemented in the multiagent system based on the developed algorithms. The algorithms were integrated in the ASTAIR’s validation platform (D4.1) and will be evaluated during the project’s validation phase (WP5).

Click below to download it!