Speaker
Description
The rapid growth of active satellites and the increasing density of orbital traffic demand a new generation of autonomous systems capable of supporting safe and efficient space operations. Within this context, the Pilot Use and Expansion of Decision Support and Coordination Systems activity advances two key technological components of ESA’s CREAM cornerstone: Automated Collision Avoidance (AutoCA) and Autonomous Space Traffic Management (AutoSTM). Together, these systems form the core of a fully integrated framework designed to automate conjunction analysis, streamline operator coordination, and enhance trust in data exchange mechanisms across the wider space traffic management ecosystem.
1. Automated Collision Avoidance (CREAM#1: AutoCA): a standalone, operator‑side decision support tool capable of autonomously assessing collision risk and recommending optimal avoidance strategies. Its architecture combines a configurable web‑based GUI, a workflow‑oriented API layer, and a computational backend responsible for ingestion, analysis, and manoeuvre design. AutoCA evaluates incoming Conjunction Data Messages, groups alerts into events, and computes collision risk using deterministic methods combined with AI/ML models trained on historical CDM datasets. These models enable prediction of future CDM evolution, improving responsiveness in dynamically evolving events and supporting more informed decision-making. The tool incorporates additional functionalities such as data fusion, uncertainty evaluation, and safety envelope determination, thereby improving the reliability of collision predictions.
2. Autonomous Space Traffic Management (CREAM#3: AutoSTM) complements AutoCA by addressing the coordination challenges that emerge during active‑vs‑active conjunction events. It provides a centralised platform where satellite operators, space surveillance providers, and mediators can exchange manoeuvre intentions and supporting data using CCSDS‑compliant standards. Building on rule‑based logic and multi‑agent negotiation principles, AutoSTM automates the filtering of events, identification of responsible parties, evaluation of proposed CAMs, and execution of negotiation or mediation procedures when required.
The combined deployment of AutoCA and AutoSTM provides significant benefits for operators and service providers. AutoCA reflects the need for late decision‑making, rapid evaluation of manoeuvre feasibility, and minimisation of false alerts reduces the need for continuous expert monitoring and accelerates the identification of avoidance manoeuvres, supporting 24/7 operations with less manpower, increasing robustness in collision avoidance operations. AutoSTM, in turn, addresses one of the most resource‑intensive aspects of space traffic management: coordinating manoeuvre responsibilities in shared orbital environments, by guiding actors through structured decision pathways, enforcing harmonised rules, and supporting complex multi‑party interactions. Their integration enables an end‑to‑end workflow—from risk assessment to coordinated mitigation—while ensuring that sensitive orbital data can be exchanged securely. Both systems are designed for iterative improvement within pilot environments, using real and simulated mission data, user feedback, and operator‑driven shadow operations to refine algorithms, interfaces, and decision logic. Complex validation and public demonstration will be performed by executing shadow avoidance and coordination operations, with a final public hosting of the system for satellite operations.
By advancing automation, data fusion, machine‑learning‑enabled prediction, and secure coordination, these systems collectively elevate the maturity of Europe’s space traffic management infrastructure. Their deployment represents a decisive step toward scalable, operator‑trusted autonomy in orbital safety processes, ensuring that future congested environments can be managed with higher reliability, transparency, and efficiency.