Challenging Autonomous Spacecraft through Trajectory Optimization with Robustness

Informacje na temat projektu

CASTOR

Identyfikator umowy o grant: 101103826

DOI

10.3030/101103826

Projekt został zamknięty

Data podpisania przez KE 26 Kwietnia 2023

Data rozpoczęcia 1 Listopada 2023

Data zakończenia 28 Lutego 2026

Finansowanie w ramach

• Marie Skłodowska-Curie Actions (MSCA)

Koszt całkowity

Brak danych

Wkład UE

€ 205 524,48

Koordynowany przez

POLITECNICO DI MILANO
Italy

Periodic Reporting for period 2 - CASTOR (Challenging Autonomous Spacecraft through Trajectory Optimization with Robustness)

Okres sprawozdawczy: 2025-03-01 do 2026-02-28

Podsumowanie kontekstu i ogólnych celów projektu

The CASTOR project (Challenging Autonomous Spacecraft through Trajectory Optimization with Robustness) emerges in response to a booming space economy characterized by an increasing number of both commercial near‐Earth and deep-space missions. Traditional ground-based Guidance, Navigation, and Control (GNC) operations are not scalable, particularly for small-scale platforms such as CubeSats, whose low-cost nature and inherent hardware limitations demand a rethinking of on-board autonomy. CASTOR aims to develop a robust, computationally efficient framework for autonomous guidance and control. Its objectives are to design an innovative closed-loop guidance algorithm, capable of re-optimizing trajectories under uncertain dynamic and environmental conditions, deploy it on spacecraft-compatible hardware, and validate the complete system in a laboratory setting simulating close-proximity operations around minor bodies. In doing so, the project addresses critical challenges like reducing human intervention and operational costs while ensuring high-performance autonomous space missions.

Prace wykonane od początku projektu do końca okresu sprawozdawczego oraz najważniejsze dotychczasowe rezultaty

During the project, the expected objectives were reached. The core algorithm, termed COLAG (Closed-Loop Autonomous Guidance), has been developed by integrating sequential convex programming for efficient trajectory optimization with a state-of-the-art uncertainty propagation method based on polynomial chaos expansion. This novel two-layer approach has enabled the algorithm to compute fuel-optimal trajectories and satisfy scientific constraints with a high degree of confidence, even under the influence of significant uncertainties in dynamics and state observations. The algorithm has been tested in scenarios, mimicking realistic space missions about minor bodies. Then, the Autonomous Guidance Computer, represented by the deployment of the algorithm on a System-on-a-chip employing both a CPU and a GPU, has prototyped and tested. This validation step ensured that the guidance method executes within the prescribed time frame using onboard computing resources. Comprehensive testing in a simulated close-proximity environment has been performed in the returning phase to prove the robustness and reliability of the system, paving the way for further integration steps and potential in-orbit demonstrations.

Innowacyjność oraz oczekiwany potencjalny wpływ (w tym dotychczasowe znaczenie społeczno-gospodarcze i szersze implikacje społeczne projektu)

CASTOR pushes the envelope compared to traditional GNC methods by enabling fully autonomous onboard trajectory optimization in highly perturbed environments. Unlike classical approaches that rely on extensive ground support and lengthy computation times, the COLAG algorithm within the CASTOR project achieves near real-time re-optimization, thereby dramatically reducing the turnaround time from days to minutes. This breakthrough is accomplished by successfully combining advanced optimization algorithms with novel uncertainty quantification techniques and an innovative hardware-software co-design approach. The method not only optimizes performance under nominal conditions but also provides resilience against deviations and unforeseen perturbations. This represents a major advancement in the field of autonomous spacecraft operations and has the potential to lower mission costs, reduce reliance on large ground teams, and facilitate more ambitious deep-space exploration initiatives.