Slide1: Terra (MODIS) Transforming Space Missions Into Service Oriented Architectures Dan Mandl NASA/GSFC Stuart Frye MitreTek Pat Cappalaere Vightel September 12, 2006 EO-1 (ALI & Hyperion) Aqua (MODIS) MODIS Active Fire Map Sensor Planning Services (SPS) 18th Global Grid Forum:


Vision: Sensor Web Enablement via a Service Oriented Architecture (SOA): Vision: Sensor Web Enablement via a Service Oriented Architecture (SOA) Land Remote Sensing Observation Data Scientists Earth Weather Data Space Weather Data Sensor Planning Services (SPS) Sensor Alert Services (SAS) Sensor Registry Services (SRS) Sensor Observation Services (SOS) Work Flow Chaining Services (WfCS) Abstract data from process of obtaining data via services above Access sensors and data via Internet and use services similarly to how “Google Earth” is used User chains multiple services from various sensors and data service providers together as needed Built on top of GMSEC and cFE


Service Example: Service Example Deliver Package (DP) User discovers service on Internet with search tools User picks desired service, pays and doesn’t get involved in details of how service is provided New services can be easily plugged in and removed thus circumventing risk of obsolescence Fault tolerant because user can locate and connect to alternative service Front end of service “Generic request” Back end “plug-ins” to fulfill service UPS FedEx Airborne Next Day $14.95 2 Day $7.95 Slow boat to China $1.25 UPS FedEx Airborne Service Aggregator “I want a package delivered”


SPS Example: Discovering and Tasking EO-1 Sensors (OGC OWS-4 Demo): SPS Example: Discovering and Tasking EO-1 Sensors (OGC OWS-4 Demo) Sensor Planning Service (SPS) Front end of service “Generic request” Back end “plug-in” to fulfill service EO-1 Service Aggregator EO-1 Tasking Map ST5 S/C ABC SAS SPS SOS Future Expansion SensorML Wrapper EO-1 Internet CASPER Onboard Planner Store capabilities Process user query and return the result Accept user goal requests Automatically sort by priority Perform auto-sync with onboard planner- CASPER Auto-sync Wrap EO-1 satellite in SensorML and publish its capabilities Enable generic command /tasking request via SPS Enable generic alert services via SAS


SPS Example: Discovering and Tasking EO-1 Sensors (OGC OWS-4 Demo): SPS Example: Discovering and Tasking EO-1 Sensors (OGC OWS-4 Demo)


SPS Example Using ST5 Built on Top of GMSEC: SPS Example Using ST5 Built on Top of GMSEC Sensor Planning Service (SPS) Validated new “back-end” predictive models which predicted problems for selected subsystems (SSR, RF Link, Power) and then autonomously initiated corrective actions through planning system before problem occurred - Unique innovation--Models self-update over time using real-time telemetry (e.g. as solar array degrades, charge current for battery changes over time, therefore model of state of battery has to change) Used GSFC Mission Services Evolution Center (GMSEC) message bus to enable communications between support components Front end of service “Generic request” Back end “plug-in” to fulfill service Land Remote Earth Weather Space Weather ST5 Service Aggregator Sensor Types S/C XYZ S/C ABC Power SSR RF Link Future Expansion Support Apps Interoperable Message Bus ST5 Planner ST5 T&C Matlab / Simulink SSR Power RF SimulinkST5 Future GMSEC Bus


Example of Service Chain to Fulfill Science Data Processing Needs: Example of Service Chain to Fulfill Science Data Processing Needs Sensor Planning Services (SPS) Sensor Alert Services (SAS) Sensor Registry Services (SRS) Sensor Observation Services (SOS) Work Flow Chaining Services (WfCS)—e.g BPEL Web Processing Service (WPS) Web Coverage Service (WCS) Web Feature Services (WFS) WebMap Services (WMS) Subset raw data by band or time Process data and georectify- e.g.level1g, also Classify--Large!! Subset by geographic area or time of acquisition Web browse image Filter by feature, area, time (Discovery by feature) Put on Map background Smart Registry - eBRIM Find (Discover) sensors that can image wildfire


Advantages of SOA for Space Sensors: Advantages of SOA for Space Sensors Networked standardized interface connections, loosely coupled Components connected at run-time Enables discovery of services Hides details of how service performed (encapsulated implementation) Fault tolerant Since connection occurs at run-time, if service not available, a component can find or “discover” an alternative service and if unavailable, can connect to another instance of the service if available Troubleshooting is easier because information is provided at component and services level Highly reusable Standardized, networked “plug and play” interfaces Scalable Interactions between services and clients independent of location and numbers Sustaining engineering for constellation simplified Can initiate new instance of service or alternative service and then disconnect old services Taken from: Hartman, Hoebel; “Lightweight Service Architectures for Space Missions”, SMC-IT 2006, Pasadena, Ca


Slide9: Key Capabilities Implemented to Enable EO-1 Sensor Webs & Support “Backend” of SPS


Slide10: CASPER Planner – response in 10s of minutes SCL – response in seconds with rules, scripts EO 1 Conventional Flight Software reflexive response ASE Flight Software Architecture Onboard Science Band Extraction Observation Planner Spacecraft Hardware Raw Instrument Data Image Overflight Times High level S/C State Information Plans of Activities (high level) Sensor Telemetry Commands (low level) S/C State Control Signals (very low level) Observation Goals L2 – Model-based Mode Identification & Reconfiguration Autonomous Sciencecraft Conventional Systems


Slide11: Original Operations Flow to Task Sensors & Access Science Data GSFC USGS Engineer MOPSS CMS ASIST Science Processing telemetry commands FDSS White Sands station in-views Mission Planner tracking data overflights contacts raw science data targets, engineering requests targets weekly schedule selected weekly schedule daily activities daily commands processed science data


Slide12: Revised Operations Flow To Task Sensors and Access Science Data Using Onboard Autonomy WWW GSFC USGS JPL ASPEN FDSS White Sands ASIST raw science data tracking data goals telemetry overflights weekly goals targets, engineering requests targets targets daily goals Science Processing EO-1 CASPER Science Processing SCL activities commands science data goals station in-views contacts processed science data Onboard EO-1 Note that engineer and mission planner removed SPS SOA Users targets


Key Differences: Key Differences Scenes and ground contacts are selected automatically based on scene priorities World Wide Web interface for requesting and acquiring observations High-level scene and contact “goals” are uploaded to the spacecraft instead of detailed command sequences Execution sequence can be automatically changed on-board Priority observations can be requested and acquired within hours Science data is immediately available for analysis on-board compared to days or weeks


Slide14: Various EO-1 Sensor Web Experiments Conducted


Prototype Workflow Chain Service (WfCS) to Enable EO-1 Sensor Web to Targets National Priority Wildfires: Prototype Workflow Chain Service (WfCS) to Enable EO-1 Sensor Web to Targets National Priority Wildfires 1 Identify NIFC-tracked Wildfire Incidents Aqua or Terra MODIS data GSFC’s Science Goal Monitor 2 Fire location confirmed and selected for imaging 4 UMD Natural Hazards Investigation Team 5 3 3 Active Fires Detection Map Roberts Fire Roberts Fire 1 Roberts Fire USFS Burned Area Emergency Response (BAER) team 6 SGM Correlate latest fire location information with MODIS imagery SGM adds target to EO-1 ground & on-orbit planning & scheduling systems and tasks EO-1 L1 Data


Slide16: MODIS Rapid Response Active Fire Detections EO-1 Advanced Land Imager Burn Scar Image On 11-2-03, the NASA Wildfire SensorWeb was employed to collect data on the burn scars resulting from the Simi Valley, Val Verde and Piru fires in Southern California. MODIS active fire detections for the duration of the event were used to target an acquisition by the ALI and Hyperion instruments onboard EO-1. Such data are employed by the USDA Forest Service for Burned Area Emergency Rehabilitation mapping. BAER maps are used to target high risk areas for erosion control treatments. In this image, burned areas appear red while the unburned areas appear green. The blue burn perimeter vector is based on ground data. Example of Rapid Delivery of Information for Decisions for EO-1 Sensor Web with WfCS


Slide17: Science Agents Science Event Manager Processes alerts and Prioritizes response observations ASPEN Schedules observations on EO-1 EO-1 Flight Dynamics Tracks, orbit, overflights, momentum management Science Alerts Observation Requests Updates to onboard plan Science Campaigns Scientists Integrated Flow forSensor Planning Service for EO-1: First Step


Beginning to Implement Standards: Beginning to Implement Standards GSFC Mission Systems Evolution Center (GMSEC) Core Flight Executive (cFE) Core Flight System (CFS) SensorML OGC Sensor Web Enablement (SWE) standards Distributed Multi-Protocol Message Bus GMSEC For Ground cFE For Space Interoperability and Meta-Language Services SensorML IRC Others TBD Application Services SensorML Archives Event Processing Create Integrate Deploy Manage Integrated Services Others TBD Instruments Satellites/ Sensors Rovers Ground Sensors


GMSEC Extended to S/C Bus--Onboard Integrated Message Bus Demonstration (December 2005): GMSEC Extended to S/C Bus--Onboard Integrated Message Bus Demonstration (December 2005) GMSEC Bus Ground System Testbed Core Flight Executive (cFE) on CHIPS UDP DC – Data Center ASIST – Advanced Spacecraft Integration and System Test cFE Bus model script result


Moving ST5 Models Onboard CHIPS Satellite Under cFS to Demonstrate Mobile Agents: Fairmont, WV Moving ST5 Models Onboard CHIPS Satellite Under cFS to Demonstrate Mobile Agents ST-5 Constellation DC – Data Center ASIST – Advanced Spacecraft Integration and System Test Adapter CHIPS Satellite with cFE Bus Onboard CHIPS – Cosmic Hot Interstellar Plasma Spectrometer ST-5 – Space Technology 5 cFE Bus Demo App Goddard Space Flight Center DC GMSEC Apps AMPS ASIST GMSEC Bus via Berkeley & Wallops Ground Stations (UDP) via DSN & McMurdo Ground Stations via TCP/IP Mobile agent –autonomous software module that can easily be moved around a network Models transformed into mobile agents Worked with Solid State Recorder agent (model) first Adapter built to make compatible with both GMSEC and Core Flight Executive (cFE) Demonstrated capability to move software running on GMSEC onboard to run under cFE Demonstrates beginning step to transform missions from central control to distributed control via self-managing software


Extended Efforts: Extended Efforts GMSEC to used on SDO, GLAST, LRO GMSEC providing framework for C3I work in the Constellation Program Will be used for ground and constellation of laboratories Two recent follow-on 3 year awards from AIST ESTO call for proposal to extend ST5 efforts An Inter-operable Sensor Architecture to Facilitate Sensor Webs in Pursuit of GEOSS Key topic – Interoperability and demonstration of service oriented architecture for space missions and sensor webs PI: Dan Mandl - 3 year effort Using Intelligent Agents to Form a Sensor Web for Autonomous Mission Operations Key topic distributed mission control Extend effort depicted on slide 16 in which ST-5 components turned into mobile agents for use onboard spacecrafts with GMSEC/CFS PI: Ken Witt/ISR Co-I Dan Mandl/GSFC – 3 year effort


Extended Efforts: Extended Efforts Goddard Institute for Systems, Software and Technology Research (GISSTR) contract effort being applied to extend ST5 effort by Institute of Scientific Research (ISR) Building GMSEC compliance tester for new components Help to synergize other ESTO awards with above mentioned awards Integrate ROME in collaboration with Capitol College into TRMM, GLAST and MMS West Virginia Challenge Grant (set-aside) to be applied to develop Sensor Modeling Language (SensorML) schemas for follow-on SOA efforts SensorML schemas will describe sensor capabilities and once put in online registries, will enable discovering of those capabilities on the Internet Open Geospatial Consortium (OGC) ongoing testbed effort OGC Web Services 4 (OWS-4) June 2006- December 2006 200 member organization of OGC 40 organization participating in OWS-4 Sensor Planning Service (SPS) one of key services being demonstrated with EO-1


Acknowledgements: Acknowledgements Additional info at: eo1.gsfc.nasa.gov and ase.jpl.nasa.gov Other Team Members: GSFC: Jerry Hengemihle, Scott Walling & Bruce Trout (Microtel), Jeff D’Agostino (Hammers), Seth Shulman (Operations Lead, Honeywell), Lawrence Ong (SSAI), Stephen Ungar (EO-1 Scientist), Thomas Brakke JPL: Steve Chien (ASE PrincipaI Investigator), Rob Sherwood (ASE Experiment Mgr), Daniel Tran (Software Lead), Rebecca Castano (Onboard Science Lead), Ashley Davies (Science Lead), Gregg Rabideau, Ben Cichy, Nghia Tang Interface & Control Systems (ICS): Darrell Boyer, Jim Van Gaasbeck Univ. of Maryland: Rob Sohlberg (Wildfire Sensor Web) Univ. of AZ: Victor Baker, Felipe Ip, James Dohm ASU: Ronald Greeley, Thomas Doggett MIT LL: Michael Griffin, Hsiao-hua Burke, Carolyn Upham


Acronyms: Acronyms ASE – Autonomous Sciencecraft Experiment ASIST – Advanced Spacecraft Integration and System Testing ASPEN – Automated Scheduling Planning Environment CASPER – Continuous Activity Scheduling Planning Execution and Replanning CMS – Command Management Systems MOPSS – Mission Operations Planning and Scheduling Systems SCL - Spacecraft Command Language


Backup: Backup


Underlying “Plug and Play” Message Bus Architecture-- Goddard Mission Services Evolution Center (GMSEC): Underlying “Plug and Play” Message Bus Architecture-- Goddard Mission Services Evolution Center (GMSEC) GMSEC architecture provides a scalable and extensible ground and flight system approach Standardized messages formats Plug-and-play components Publish/Subscribe protocol Platform transparency ST5 first mission to be totally GMSEC compliant More info at: http://gmsec.gsfc.nasa.gov


Core Flight Executive (cFE), an Extension for GMSEC for Flight SW: Core Flight Executive (cFE), an Extension for GMSEC for Flight SW cFE provides a framework that simplifies the development and integration of applications Layered Architecture – software of a layer can be changed without affecting the software of other layers Components communicate over a standard message-oriented software bus, therefore, eliminating the need to know the details of the lower layers of inter-networking. Software components can be developed and reused from mission to mission. Developed by Flight SW Branch at GSFC To be used on LRO More info at: http://gmsec.gsfc.nasa.gov


Example of Rapid Mission Configuration Using GMSEC Interoperable Catalog Components: GMSEC approach gives users choices for the components in their system. The TRMM mission rapidly selected key components from the GMSEC catalog. Example of Rapid Mission Configuration Using GMSEC Interoperable Catalog Components