Projecte llegit

Títol: Distributed Robotic Autonomous Control for Observatory Systems (DRACOS)

---

Estudiants que han llegit aquest projecte:

DOMENE FERNÁNDEZ, FRANCISCO (data lectura: 15-09-2025)
   Cerca aquest projecte a Bibliotècnica

---

Director/a: GUTIÉRREZ CABELLO, JORDI

Departament: FIS

Títol: Distributed Robotic Autonomous Control for Observatory Systems (DRACOS)

Data inici oferta: 03-02-2025     Data finalització oferta: 03-10-2025

---

Estudis d'assignació del projecte:

MU AEROSPACE S&T 21

Tipus: Individual

Lloc de realització: Fora UPC

Supervisor/a extern: Kike Herrero

Institució/Empresa: IEEC

Titulació del Director/a: Dr en Ciències Físiques

Paraules clau:

Robotic Astronomy, Autonomous Observatories, Observatory Control System (OCS), Robot Operating System 2 (ROS 2), Behavior Trees, Distributed Control Systems, Real-Time Systems, Systems Architecture, Lifecycle Management, Time-Domain Astrono

Descripció del contingut i pla d'activitats:

The rapid evolution of robotic observatories creates a need for a standardized framework that can seamlessly integrate and control the diverse array of devices and systems commonly found in observatories, such as telescopes, instruments, domes, and other hardware. This project aims to develop a next-generation framework built on ROS2 (Robot Operating System 2) to address this need. The proposed framework seeks to set a new standard for observatory automation and operation by prioritizing modularity, interoperability, and scalability, enabling communication and coordination across all components. The system will feature an adaptive and autonomous architecture capable of managing complex workflows and scenarios without human intervention. Advanced algorithms and state-of-the-art control methodologies will be implemented to handle intricate dependencies, ensuring smooth operation even under dynamically changing conditions. The framework's modular design will allow developers and observatory operators to integrate new devices or functionalities, enabling long-term sustainability and reducing maintenance efforts. As a pilot system, this framework will be tested and validated at the Observatori del Montsec (OdM) with its newest telescope, TRAC. The project will leverage OdM's state-of-the-art infrastructure to demonstrate the framework's capability to coordinate diverse subsystems in a real-world observatory environment. The capacity for autonomous decision-making and coordination across all observatory nodes facilitates robust and reliable operation through sophisticated state management. With a focus on achieving true standardization, the system will enhance compatibility between diverse hardware and software components, empowering observatories worldwide to adopt a unified operational paradigm. This project aspires to revolutionize the field of robotic and autonomous observatories by delivering an adaptable, efficient, and standardized solution that advances astronomical research and reduces the operational complexity of observatory management.

Overview (resum en anglès):

The rapid growth of time-domain sky surveys has made it increasingly important to have robotic observatories that can operate reliably without human intervention. Yet, much of today's obser- vatory control software still depends on fragmented, vendor-specific tools and often lacks the strong architectural foundations needed to handle the challenges of fully unattended operation, particularly in terms of concurrency, safety, and modularity. This thesis addresses these gaps by introducing the Distributed Robotic Autonomous Control for Observatory Systems (DRACOS), a unified control framework designed to deliver a robust and extensible solution for autonomous observatory management. DRACOS is built on the Robot Op- erating System 2 (ROS 2), applying modern robotics principles to astronomy in a systematic way. At its core is a hierarchical, agent-based architecture that combines a clear chain of command with a distributed network of modular software agents. A central innovation is the Subsystem Agent Contract, which requires all hardware interfaces to be implemented as ROS 2 Lifecycle Nodes ensuring deterministic, system-wide state management that is essential for dependable autonomous operations. Instead of relying on traditional procedural scripts, DRACOS uses Behavior Trees to model com- plex, reactive tasks. This approach integrates error handling and recovery directly into task exe- cution and is backed by a high-priority, event-driven safety system capable of instantly halting all activities in response to critical events. The framework also includes a web-based user interface and has been validated through a test campaign, using both industry-standard simulators and real-world on-sky trials with the new 0.4 m TRAC robotic telescope at the Montsec Observatory. Results show that DRACOS meets its requirements for reliability, safety, and performance, successfully achieving fully autonomous op- eration. Overall, this work positions DRACOS as both a practical solution for today's observatories and an architectural blueprint for the next generation of robotic astronomy facilities.

---