Space activities, such as On-Orbit Servicing (OOS) missions, are currently experiencing a deep transformation thanks to the advances in space robotics. Refueling, maintenance and many other services could be provided in future by suitable spacecraft equipped with robotic arms able to safely grasp a client satellite. However, these types of missions pose several technical challenges never faced until now. Of utmost relevance is the modelling and control of such a complex dynamical system (i.e. relative motion, robotic arms control, appendages' flexibility and sloshing phenomena), needed to effectively achieve the mission objectives.
Furthermore, multidisciplinary field such as automation, guidance and navigation, multibody and structural dynamics are involved in the design of the various phases of an OOS mission. As a result, this work will only focus on the last steps of a refueling mission and the major contribution will be given in the modelling of flexible elements (i.e. solar array, robotic arms and joints) and sloshing dynamics. Nevertheless, while the modelling of flexible appendages is nowadays common practice via the well-known modal decomposition technique, no unique model exists to accurately represent fuel sloshing in every scenario except for Computational Fluid Dynamics (CFD), which is however impractical for control design purposes. To address such issue, several Equivalent Mechanical Models (EMM) will be implemented to characterize the fluid dynamics and variation in the fuel mass during refueling will be opportunely accounted with time-varying parameters in the EMM.
The resulting dynamical model will be then used both for control synthesis and for verification and validation purposes. Structured Hinf and µ-synthesis methods will be used for the generation of the robust controllers required for the capture phase and the post-grasping operation. Finally, µ-analysis and non-linear Monte Carlo tests will be presented to check the required performances.
The proposed research project has as goal the evaluation of the disturbing effect produced by flexible dynamics and sloshing phenomena on the performance of the chaser AOCS and the robotic arms manipulation during the last phases of a refueling mission. This research could potentially be of great interest for space agencies and industries which are currently pushing towards robotic on-orbit operations for servicing, repairing or de-orbiting.
The areas in which the proposed research activity could potentially give a relevant contribution are listed here:
- The inclusion of a high-fidelity description of the flexible dynamics directly in the linear synthesis process of the controller. In most cases, instead, only simple models of the structural system are taken into consideration at this stage and increased uncertain ranges in the parameters are used to account for that simplification. The latter approach could potentially produce some over-conservative results from the control synthesis.
- While the technical community has studied the effect of sloshing during attitude control and mission specific data regarding fluid slosh in microgravity have been produced, the modelling of this phenomena for the study of grasping and refueling operation is currently not thoroughly investigated. In this research the interaction of the fluid slosh with the robotic manipulator control and its effect on the overall dynamical system during the contact phase between the two spacecraft will be investigated with Equivalent Mechanical Models (EMM). Furthermore, the researcher will evaluate how to translate the variation of the fuel mass during refueling in time variation of the parameters of the EMM.
- The application of new optimization strategies for computing lower and upper bound of µ manifests some problems whenever lightly damped system with very large Linear Fractional Transformation (LFT) models [1] are considered. Since very few dynamical systems present a size of the state-space model comparable with flexible spacecraft equipped with a robotic manipulator, this research could constitute a valuable benchmark for the evaluation of the effectiveness of optimization algorithm in determining a good estimate of the structured singular value [2].
- The improvement of an in-house open-source code for the implementation of the control design procedures and an interface module between the Finite Element Modelling (FEM) commercial software Patran/Nastran and Matlab for exporting the FE structural model and the propagation of uncertainties.
[1] A. Kron et al., Enhanced Linear Fractional Transformation: a Matlab Toolbox for Space System Modeling and Controller Analysis and synthesis. IFAC Symposium on Automatic Control in Aerospace (2013).
[2] G. Ferreres et al., Robustness analysis of flexible structures: practical algorithms. International Journal of Robust and Nonlinear Control, Vol. 13 (2003), 715-733.