Deploying Interactive Mission Planning Tools- Experiences and Lessons Learned -

Abstract

This article contains a retrospective overview of connected work performed for the European Space Agency (ESA) over a span of 10 years. We have been creating and refining an AI approach to problem solving and have infused a series of deployed planning and scheduling systems which have innovated the agency’s mission planning practice. The goal of this paper is to identify strong features of this experience, comment on general lessons learned and offer guidelines for work practice of the future. Specifically, the work considers some key points that have contributed to strengthening the effectiveness of our approach for the development of an end-to-end methodology to field applications: the attention to domain modeling, the constraint-based algorithm synthesis and the relevance of user interaction services.

Cluster-II: Using Artificial Intelligence for Automated Ground Station Scheduling

“…Esa contributed in this area by promoting the development of Apsi and Apsi-related activities for on-ground AI application deployment. 9,10 Tiago's underlying planning technology is based on constraint based temporal planning with timelines. The planner is deployed using two classes of Apsi modelling components: state variables and resources (for a more detailed description of the modelling and problem solving behind Tiago see Ref.…”