Insights from IWPSS 2025 in Toulouse
The Earth Observation (EO) landscape is undergoing a tectonic shift. As demand for timely, accurate and cost-effective satellite imagery increases, traditional monolithic mission planning models and siloed ground segment operations are proving inadequate. New concepts such as Ground Station as a Service (GSaaS) and federated satellite networks necessitate sophisticated coordination, communication and responsiveness models across diverse EO systems.
This evolution was at the forefront of discussions at the 2025 International Workshop on Planning and Scheduling for Space (IWPSS), held in Toulouse, France. The IWPSS is the leading conference for research into decision-making, autonomy and temporal reasoning in aerospace systems. The 2025 edition brought together leading researchers and practitioners specialising in autonomous planning for satellite operations, with a particular focus on federated architectures and AI-enhanced scheduling systems.
Two papers from the DOMINO-E consortium stood out for their complementary approaches to the shared challenge of coordinating heterogeneous Earth observation (EO) systems in a dynamic and uncertain environment. One paper addressed the issue of communication scheduling across ground segments, while the other tackled area coverage via distributed mesh dispatching.
Toward Federated Autonomy in EO Mission Planning
Both contributions focus on a key operational bottleneck: how to schedule and allocate Earth observation (EO) resources across multiple satellite constellations and ground systems, while dealing with uncertainty, scale and conflicting objectives.
The first paper, titled ‘Deterministic and Probabilistic Decision Models for GSaaS-Based Satellite Communication Resource Management‘ and authored by Willot, Farges, Picard and Pavero, presents a dual modelling framework for scheduling data transfers between Low Earth Orbit (LEO) satellites and a network of ground stations. This framework innovatively combines deterministic optimisation with a probabilistic reasoning layer, providing a more accurate reflection of the inherent variability in orbital dynamics and GSaaS slot availability.
Our findings contribute significantly to the field, offering a pathway to enhanced resource management and data exchange efficiency in the evolving landscape of EO satellite communications.” (p.1)
In this model, the Satellite Communication and Resource Management System (SCRMS) acts as an intermediary for private and shared ground station access. The SCRMS integrates predictions of orbital access windows with booking constraints and prioritisation logic to ensure that communication slots are dynamically reallocated as conditions change.
Coverage Scheduling Across Federated Systems
Meanwhile, the second paper, ‘Mesh Dispatching for Area Coverage Using Several Earth Observation Systems‘ by Pralet, Picard, de Lussy and Guerra, focuses on the dispatching of observation tasks across multiple EO missions. The paper introduces a federated service that divides large geographic requests into smaller meshes and allocates them to available missions, each of which has its own independent planning engines and operational constraints.
“The dispatch strategy is optimized based on heuristic search and large neighborhood search… re-dispatch operations are also used to re-optimize the allocation of meshes to systems depending on the current progress.” (p.1)
By combining heuristic techniques with Large Neighborhood Search (LNS), the framework can handle scenarios such as capturing satellite images of an entire country (e.g. France), adapting in real time to weather or mission load failures. This mirrors the challenges tackled by SCRMS, but instead of bandwidth allocation, the focus is on observation capacity.
A Shared Challenge: Dynamic Scheduling in the Face of Heterogeneity
Both contributions highlight a key theme: the importance of dynamic, cross-entity coordination for large-scale Earth observation. The complexities are similar, whether the resource is communication bandwidth or imaging time.
- Multiple autonomous actors (e.g. satellite owners, GSaaS providers, mission centres).
- Diverse constraints (e.g. orbital geometry, weather, antenna visibility, cost models).
- Continuous uncertainty and the need for re-optimisation.
This aligns with the broader goals of the DOMINO-E project, which advocates a modular, federated EO ground segment. The project introduces reusable, interoperable software ‘Dominoes’ for services such as coverage planning, communication booking and user access. These are built to handle the distributed and dynamic nature of future EO infrastructures.
Toward Operational Scalability in the New Space Ecosystem
Together, the contributions to the IWPSS 2025 signal a clear trajectory for Earth observation (EO) system architecture, moving from isolated mission planning to intelligent, federated orchestration. As GSaaS and commercial satellite constellations become more widespread, the ability to dynamically allocate resources, revise schedules at short notice and balance priorities between stakeholders will be crucial for mission success. In this respect, both the SCRMS and the Mesh Dispatcher are not just algorithms; they are critical enablers for the future of responsive Earth observation.
Sources
Hénoik Willot et al. 2025. Deterministic and Probabilistic Decision Models for GSaaS-based Satellite Communication Resource Management. Conference Paper presented at the International Workshop on Planning & Scheduling for Space. Toulouse, April 28-30, 2025. Url: https://domino-e.eu/wp-content/uploads/2025/05/15_Deterministic_and_Probabili.pdf
Pralet, Cédric et al. 2025. Mesh Dispatching for Area Coverage using Several Earth Observation Systems. Conference Paper presented at the International Workshop on Planning & Scheduling for Space. Toulouse, April 28-30, 2025. Url: https://domino-e.eu/wp-content/uploads/2025/05/21_Mesh_Dispatching_for_Area_C.pdf