Many large airports aim to have complete autonomous airside operations in the future. Amsterdam Airport Schiphol (AAS) for example, launched the Autonomous Airside Operations program to achieve this goal. Our main contribution is to present a Multi-agent Pickup-and-Delivery (MAPD) model that uses a centralized task allocation mechanism to improve the performance of integrated task allocation and path planning for autonomous ground handling operations compared to previous research. This study models a global multi-vehicle Pickup and Delivery Problem with Time Windows (PDPTW) for the scheduling of autonomous ground handling tasks. A warm start multi-objective mixed integer linear programming model is proposed to solve the scheduling problem where the initial feasible solution is obtained by an insertion heuristic. This multi-agent task allocation model, when combined with multi-agent path planning, forms a MAPD model for modeling autonomous ground handling operations. Multi-agent path planning is solved using prioritized Safe Interval Path Planning (SIPP). A replanning model is developed to assess the resilience of our model to disruptions of operations. Also, a mixed integer nonlinear programming model, which includes an additional non-linear objective, is proposed to generate more realistic task assignments by minimizing the waiting time of vehicles on the aircraft stands. In this study, a four-hour planning window with three aircraft stands at AAS is used for the experiments. The results show that the proposed approach improves the computational time of the task allocation model with 48% for the normal traffic scenario, compared to the previously published results. The conflict-free routes of all ground support equipment (GSE) vehicles are all successful and close to the shortest path results, with an average increase of 0.04% and 10% for the path length and the duration of the path, respectively. Our model is therefore able to generate complete, high quality solutions in less than three minutes.
... Read more

Inspired by the vision of fully autonomous airside operations at Schiphol airport, this study aims to contribute to the short-term goal of automated aircraft ground handling. In this research, we design and evaluate a multi-agent system for planning of automated ground handling. There are two main components in the system, task allocation optimization, and multi-agent path planning. To allocate tasks to ground support equipment (GSE) vehicles, an auction mechanism inspired by temporal sequential single item (TeSSI) auction is proposed. Ground handling tasks scheduling for GSE vehicles is modeled as several single-vehicle pickup and delivery optimization problems (SPDP), and the values of the objective functions are applied to generate bids for GSE vehicle agents in the auction. Moreover, Prioritized Safe Interval Path Planning for large agents (LA-SIPP) is used to plan collision-free paths for GSE vehicle agents in the model to execute tasks. Experimental studies have shown that the system is able to perform task allocation and path planning of ground handling tasks for flights in 3 aircraft stands within a 4-hour time in a reasonable computational time. Moreover, the model is capable to replan the tasks for agents when disruption happens. Applying the lowest possible numbers of vehicles used in the current operation, the model can always reach success allocation and path planning rates higher than 81% and 98%, respectively. ... Read more