Real Time System Identification for Complex System Modelling and Autonomy Operations - Expro Plus

• Country: Netherlands
• Summary: Real Time System Identification for Complex System Modelling and Autonomy Operations - Expro Plus
• NLT Ref No: 102987852
• Deadline: 17 Jul 2024
• Financier: Agency for the Cooperation of Energy Regulators (ACER)
• Purchaser Ownership: Government
• Tender Value: Refer Document
• Notice Type: Tender
• Document Ref. No.:
• Purchaser's Detail:
  Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details. Login to see full purchaser details.

  Login to see details
• Description:
• Tenders are invited for Real Time System Identification for Complex System Modelling and Autonomy Operations - Expro Plus Objectives: To develop on-board system-identification algorithms for on-line monitoring, decision-making and analysis. Description:Future spacecraft missions require high accuracy on-board models in order to enable active monitoring, health diagnosis as well as control adaptation and reconfiguration. Accurate pre-flight knowledge of the plant to be controlled is today necessary for designingaccurate Guidance, Navigation Controls (GNCs) for stabilisation and instrument pointing systems (including for instance precise structural modal analysis). However, in autonomous operations the dynamical models may not always be available online. When operatingin situ initial on-board models may not always have the desired accuracy when derived from analytical physical modelling principlesor ground experiments. This motivates the need to use on-board system identification techniques for the generation of in situ-modeldynamics. The on-board modelling techniques addressed in this study are categorized as model-based estimation and identification techniques for the generation of dynamical models and model calibration includes machine-learning-based strategies. This activity aimsat developing system identification techniques to determine accurately in a computational efficient and reliable manner model dynamics and parameters for control purposes (control, adaptation, monitoring, estimation, perception and reconfiguration). These techniques shall be suitable for on-line identification and therefore must be capable of providing meaningful output without a pre-defined excitation input. Novel system identification techniques shall be developed and tested with the purpose to generate online high-fidelity models which cannot be reliably predicted by low order computational models (uncertainties, non-linearities, un-modelled dynamics etc..), sub-scale experiments or non-representative ground experiments (slosh in micro-g). Model validation and verification issues associated with the development of the on-board modelling strategies including uncertainty characterisation shall be included in the study. The activity encompasses the following tasks:- Review and selection of on-board uncertainty modelling techniques- Assessment of the dynamic and kinematic modelling requirements for a high performance mission- Perform a trade-off study between the various on-line model identification techniques and perform a final selection of candidate techniques- Development of a SW prototype and its' integration into a GNC functional simulator- Evaluation of the impact on system performance
• Documents:

 Tender Notice