9, December 2016

ESA Open Invitation To Tender AO8812
Open Date: 07/12/2016
Closing Date: 19/01/2017 13:00:00


Status: ISSUED
Reference Nr.: 16.117.02
Prog. Ref.: GSTP Period 6 E1 PRJ
Budget Ref.: E/0904-611 – GSTP Period 6 E1 PRJ
Special Prov.: LU+BE
Tender Type: C
Price Range: 200-500 KEURO
Products: Ground Segment / Mission Operations / Operations Execution / Configuration management, FOP tools, On-board resource checker, …
Techology Domains: Mission Operation and Ground Data Systems / Ground Data Systems / Human¿Computer Interfaces and Technologies
Establishment: ESOC
Department: Mission Operations Department
Division: Studies & Special Projects Division
Contract Officer: Hurtz, Anne Maria
Industrial Policy Measure: N/A – Not apply
Last Update Date: 07/12/2016
Update Reason: Tender issue

The objectives are to:- take the output of a Data Mining Mission History GSP study and improve the two algorithms (i.e. Kernel Density Estimation and Poincare analysis) to the point that they could be specifiedfor a future operational system.- cover reliable automatic telecommand verification checking to point that thenecessary algorithms can be specified for a future operational system.-determine the requirements and characteristics of the necessary advancedMMIs (man machine interfaces) and APIs (application programming interfaces) that would be required to allow the spacecraft operationsengineers to intuitively and efficiently interact with the large data sets involved- build an end to end proof of concept system capable of allowing spacecraft operations engineers to provide feedback on how best these algorithms (in combination with a suitable MMI)can be used to perform better spacecrafthealth analysis, telemetry checking and telecommand verification in a futureoperational system.Operations engineers check how the mission (space, ground control) is performing as expected by implementing checks in the operational databases. These are checks to see if spacecraft parameter values are between predefined limits or in predefined statuses. There are also command verification checks containedin the procedures or integrated into the control system. These are checks on spacecraft parameter values after a command has been sent.Real-time telemetry limit checks and telecommand verification checks take effort todefine, implement and manage. Two importanttrends are that the number of spacecraft telemetry parameters and telecommands keeps increasing and the manpower for operating the ground systems keeps being reduced. This means that the present systems, involving engineers defining these checks individuallyand manually in operational databases/ procedures, are incomplete. What isneeded in the future is a system that can derive checks automatically from historical data. Once such an automatic system exists it can be expanded to check many more spacecraft telemetry parameters, verify more commands and check more procedure steps with very little effort on the ground side.In 2009 ESA/ESOC deployed an automatic checking algorithm called Novelty Detection. Based on the operational success, a GSP study (Automatic Spacecraft Status Characterisation by data mining mission history) was started in 2014 to see how it could be expanded and built upon. An approach use two generic techniques based on Poincaré-Plots and on Kernel Density Estimates (KDEs) was developed. The work determined correlationsbetween commands and their impact on telemetry parameters. The study showed new ways of presenting andmanipulating the enormous amountof data generated would be required for non data analytics experts to interact with the system. It also showed that improvements inthe algorithms would be required, especially in terms of speed and usability before they could be specified for an operational deployment.This study will address those issues and will enable ESA to specify the future operational system with a high degree of confidence and accuracy thanks to an iterative test and improve cycle with the end users. As such the proof of concept shall be compatible with the present Analysis and Reporting System at ESOC. Theaim will be to reuse as much as possible from the present ESA data mining and MMI infrastructure e.g. Hadoop services and ESOC’s EGOS User Desktop Project.There will be eight tasks. Tasks 1-4 are concerned with improving the algorithms and Tasks 5-7 with the MMI. Task 8 is wrap up and specification freezing.Throughout the projectESA will be responsible for incorporating industries inputs and maintaining the proof of concept, allowing fast feed back from spacecraft engineers controlling real spacecraft. Task 1 concerns the consolidation of the GSP KDE and Poincare methods, including the algorithm specification for a first iteration on the proof of conncept i.e. algorithm implementation in ESOC. Task 5 will start in parallel, with an analysis of the suitability of ESOC MMI infrastructure and a first set of visualisation/MMIfunctionalities withmockup user validation sessions. A review of the ESA EGS-CCdata model will be made to ensure that any constraints in using these techniques for AIV are taken into account.Task 2 is then about further expanding the methods / exploring further possibilities,which would be provide inputs (algorithms, visualisation/MMI needs) to furtheriterations into the mockup/prototypes. Tasks 3 and 4 concern the applicability of the approach to telecommand verification and will follow a similar logic.Tasks 5 to 7 which are concernedwith the MMI aspects will proceed in a recursive manner as the results from Task 1-4 flow into it and based on user feedback from all parties is integrated. Task 8 will then freeze the algorithm specifications and MMI requirements. The project will then be wrapped up.

If you wish to access the documents related to the Invitation to Tender, you have to log in to the ESA Portal.