Sensor Web Enablement SWE and Sensor ML January
Sensor Web Enablement (SWE) and Sensor. ML January 2008 Mike Botts mike. botts@uah. edu Principal Research Scientist University of Alabama in Huntsville Mike Botts – January 2008 1
Mike Botts – July 2006 2
Open Geospatial Consortium • The Open Geospatial Consortium, Inc (OGC) is an international industry consortium of 334+ companies, government agencies and universities participating in a consensus process to develop publicly available interface specifications and encodings. • Open Standards development by consensus process • Interoperability Programs provide end-to-end implementation and testing before spec approval • Standard encodings (e. g. GML, Sensor. ML, O&M, etc. ) – – • Mike Botts – July 2006 Geography Markup Language (GML) – Version 3. 2 Style Layer Description language (SLD) Sensor. ML Observations and Measurement (O&M) Standard Web Service interfaces; e. g. : – Web Map Service (WMS) – Web Feature Service (WFS) – Web Coverage Service (WCS) – – Catalog Service Open Location Services – used by communications and navigation industry – Sensor Web Enablement Services (SOS, SAS, SPS) 3
Mike Botts – July 2006 4
Basic Desires • Quickly discover sensors and sensor data (secure or public) that can meet my needs – location, observables, quality, ability to task • Obtain sensor information in a standard encoding that is understandable by me and my software • Readily access sensor observations in a common manner, and in a form specific to my needs • Task sensors, when possible, to meet my specific needs • Subscribe to and receive alerts when a sensor measures a particular phenomenon Helping the World to Communicate Geographically
Sensor Web Enablement Framework Heterogeneous sensor network Airborne In-Situ monitors Surveillance - sparse - disparate Decision Support Tools Satellite Bio/Chem/Rad Detectors - mobile/in-situ - extensible Models and Simulations Sensor Web Enablement - discovery - access - tasking - alert notification web services and encodings based on Open Standards (OGC, ISO, OASIS, IEEE) - nested - national, regional, urban - adaptable - data assimilation - vendor neutral - extensive M. Botts -2004 - flexible - adaptable Helping the World to Communicate Geographically
Background -1 OGC Web Services Testbed 1. 2: OGC Web Services Testbed 1. 1: Sensor. ML initiated at University of Alabama in Huntsville: NASA AIST funding 1999 - 2000 • Sponsors: EPA, NASA, NIMA • Specs: SOS, O&M, Sensor. ML • Demo: NYC Terrorist • Sensors: weather stations, water quality 2001 • Sponsors: EPA, General Dynamics, NASA, NIMA • Specs: SOS, O&M, Sensor. ML, SPS, WNS • Demo: Terrorist, Hazardous Spill and Tornado • Sensors: weather stations, wind profiler, video, UAV, stream gauges 2002 • Specs advanced through independent R&D efforts in Germany, Australia, Canada and US • Sensor Web Work Group established • Specs: SOS, O&M, Sensor. ML, SPS, WNS, SAS • Sensors: wide variety 2003 -2004 Helping the World to Communicate Geographically
Background -2 OGC Web Services Testbed 3. 0: • Sponsors: NGA, ORNL, LMCO, BAE • Specs: SOS, O&M, Sensor. ML, SPS, Transducer. ML • Demo: Forest Fire in Western US • Sensors: weather stations, wind profiler, video, UAV, satellite SAS Interoperabilty Experiment 2005 OGC Web Services Testbed 4. 0: SWE Specifications toward approval: Sensor. ML – V 0. 0 Transducer. ML – V 0. 0 SOS – V 0. 0 SPS – V 0. 0 O&M – Best Practices SAS – Best Practices • Sponsors: NGA, NASA, ORNL, LMCO • Specs: SOS, O&M, Sensor. ML, SPS, Transducer. ML, SAS • Demo: Emergency Hospital • Sensors: weather stations, wind profiler, video, UAV, satellite 2006 OGC Web Services Testbed 5. 1 • Sponsors: NGA, NASA, • Specs: SOS, Sensor. ML, TML • Demo: Streaming JPIP of Georeferenceable Imagery; Geoporocessing Workflow • Sensors: Satellite and airborne imagery 2007 Helping the World to Communicate Geographically
SWE Specifications • Information Models and Schema – Sensor Model Language (Sensor. ML) for In-situ and Remote Sensors - Core models and schema for observation processes: support for sensor components, georegistration, response models, post measurement processing – Observations and Measurements (O&M) – Core models and schema for observations – Transducer. ML – adds system integration and multiplex streaming clusters of observations • Web Services – Sensor Observation Service - Access Observations for a sensor or sensor constellation, and optionally, the associated sensor and platform data – Sensor Alert Service – Subscribe to alerts based upon sensor observations – Sensor Planning Service – Request collection feasibility and task sensor system for desired observations – Web Notification Service –Manage message dialogue between client and Web service(s) for long duration (asynchronous) processes – Sensor Registries – Discover sensors and sensor observations Helping the World to Communicate Geographically
Status • Current specs are in various stages – Sensor. ML (and SWE Common) – Version 1. 0. 1 – Transducer. ML – Version 1. 0 – Observations & Measurement – Version 1. 0 – WNS – Request for Comments – SOS – Version 1. 0 – SPS – Version 1. 0 – SAS – Request for Comments • Approved SWE standards can be downloaded: – Specification Documents: http: //www. opengeospatial. org/standards – Specification Schema: http: //schemas. opengis. net/ Helping the World to Communicate Geographically
Known Open Source Software for SWE • 52 North / University of Muenster – Full suite of SWE services (SOS, SPS, SAS, WNS) • University of Alabama in Huntsville – – SWE Common parser/writer, Sensor. ML parser, process chain executor and model library editors for Sensor. ML/O&M instances and profiles Space Time Toolkit SWE client SOS/WCS services SWE portrayal service (initially KML) process • Texas A&M / Marine Metadata Initiative – Non eb. RIM registry based on ontology – light weight clients, several services • Map. Server / GDAL – SWE services incorporated into Map. Server • NASA GSFC / Geo. Blinky – Several components used with the EO 1 SAT activities • Northrop Grumman There is an initiative to begin to look at joint development and management of Open Source SWE software – Several components used within the Pulse. Net activity • SAIC – Ongoing development of several Open-Source SWE components under MASINT funding Helping the World to Communicate Geographically
Example Known External Activities using SWE • Community Sensor Models (NGA) – Sensor. ML encoding of the CSM; CSM likely to be the ISO 19130 standard • Multi-Int Metadata Standards (DIA) – Committed to Sensor. ML and SWE as direction • OGC OWS 5. 1 Georeferenceable Imagery (NGA/NASA) – will be demonstrating use of Sensor. ML within JPEG 2000 and JPIP for support of geolocation of streaming imagery • Oak Ridge National Labs Sensor. Net – funded project will be adding Sensor. ML support in Sensor. Net nodes for threat monitoring • Northrop Grumman IRAD (NGC TASC) – demonstrated end-to-end application of Sensor. ML/SWE for legacy surveillance sensors in field • Empire Challenge (NGA - SAIC) – planning to test SWE for sensor observation processing and integration in 2008 testbed • European Space Agency – developing Sensor. ML profiles for supporting sensor discovery and processing within the European satellite community • Canadian Geo. Connections Projects – used Sensor. ML in water monitoring network • Hurricane Missions (NASA MSFC) – using Sensor. ML for geolocation and processing of satellite and airborne sensors • Sensors Anywhere (SAny) – intending to use Sensor. ML/SWE Common within large European sensor project • NASA ESTO – funded 30 3 -year projects on Sensor Webs; 5 SBIR topics with Sensor. ML and Sensor Web called out Helping the World to Communicate Geographically
Helping the World to Communicate Geographically
What is Sensor. ML? • XML encoding for describing sensor processes – Including sensor tasking, measurement, and post-processing of observations – Detectors, actuators, sensors, etc. are modeled as processes • Open Standard – – Approved by Open Geospatial Consortium in 2007 – Supported by Open Source software (COTS development starting) • Not just a metadata language – • enables on-demand execution of algorithms Describes – Sensor Systems – Processing algorithms and workflows Mike Botts – January 2008 14
Why is Sensor. ML Important? • Importance: – Discovery of sensors and processes / plug-n-play sensors – Sensor. ML is the means by which sensors and processes make themselves and their capabilities known; describes inputs, outputs and taskable parameters – Observation lineage – Sensor. ML provides history of measurement and processing of observations; supports quality knowledge of observations – On-demand processing – Sensor. ML supports on-demand derivation of higher -level information (e. g. geolocation or products) without a priori knowledge of the sensor system – Intelligent, autonomous sensor network – Sensor. ML enables the development of taskable, adaptable sensor networks, and enables higher-level problem solving anticipated from the Semantic Web Mike Botts – January 2008 15
Sensor. ML Processes Non-Physical Processes where physical location or physical interface of the process is not important (e. g. a fast-Fourier process) Physical Processes where physical location or physical interface of the process is important (e. g. a sensor system) Atomic Processes that are considered Indivisible either by design or necessity Composite Processes that are composed of other processes connected in some logical manner Mike Botts – January 2008 16
Example Atomic Processes • Transducers (detectors, actuators, samplers, etc. ) • Spatial transforms (static and dynamic) – Vector, matrix, quaternion operators – “Sensor models” • scanners, frame cameras, SAR • polynomial models (e. g. RPC, RSM) • tie point model – Orbital models – Geospatial transformations (Map projection, datum, coordinate system) • Digital Signal Processing / image processing modules • Decimators, interpolators, synchronizers, etc. • Data readers, writers, and access services • Derivable Information (e. g. wind chill) • Human analysts Mike Botts – January 2008 17
Example Composite Processes • Sensor Systems, Platforms • Observation lineage – from tasking to measurement to processing to analysis • Executable on-demand process chains: – geolocation and orthorectification – algorithms for higher-level products • e. g. fire recognition, flood water classification, etc. – Image processing, digital signal processing • Uploadable command instructions or executable processes Mike Botts – January 2008 18
Status of Sensor. ML and SWE Common – Sensor. ML history • Influenced by interoperability challenges for satellite sensors at NASA • Started at UAH in 1998 under NASA AIST funding; brought into OGC in 2000 • Approved as Public Discussion Paper (2002) • Approved as Recommended Paper (2004) • OGC 05 -086 approved as Best Practices Document in Bonn (Nov 2005) • OGC 05 -086 r 3 approved as Version 0. 0 Technical Specification in July 2006 • OGC 07 -000 approved as Technical Specification Version 1. 0 on June 23, 2007 – Current: document (OGC 07 -000) • Approved Version 1. 0 of Sensor. ML and SWE Common data types • Official document available at OGC ( http: //www. opengeospatial. com ) • Official Reference Schema resides online at http: //schemas. opengis. net/ • Doc and schema also available at http: //vast. uah. edu/Sensor. ML Mike Botts – January 2008 19
Where has Sensor. ML been demonstrated and tested ? Mike Botts – January 2008 20
Known Demonstrations and Testbeds for Sensor. ML • Previous OGC OWS Testbeds – OWS 1. 1 (2001) – description and access to in-situ sensors – OWS 1. 2 (2002) – discovery, access, and georeferencing of remote sensors; fusion with in-situ sensors – OWS 3. 0 (2005) – discovery, on-demand processing of radar, satellite, UAV, weather station, and plume model observations – • OWS 4. 0 (2006) – discovery, on-demand processing of CBRNE, plume model, and weather sensors NASA SMART ODM (2006 -present) – have used Sensor. ML to georeference satellite data and to automatic determination of coincidence between sensor and numerical model data • Northrop Grumman IRAD (Pulse. Net 2006 -2007) – demonstrated end-to-end application of Sensor. ML/SWE for legacy surveillance sensors in field • Empire Challenge 2007 (NGA) – Sensor. ML used for discovery and data access of disparate sensor sources • Canadian Geo. Connections Projects (2005) – used Sensor. ML in water monitoring network (discovery and data access) • NASA Hurricane Missions (2006 -present) – using Sensor. ML for geolocation and processing of satellite and airborne sensors • NASA Sensor. ML Project (2006 -present) – incorporation and demonstration of Sensor. ML execution engine into Space Time Toolkit and SWE services Mike Botts – January 2008 21
Previous OGC OWS Testbeds: Sensor. ML-Enabled Discovery and Georeferencing La. Plata Tornado Weather UAV for Fire Detection Radiation plumes and weather Mike Botts – January 2008 22
A Few Known Ongoing Activities using Sensor. ML • Community Sensor Models (NGA) – Sensor. ML encoding of the CSM; CSM likely to be the ISO 19130 standard • Multi-INT Metadata Standards (DIA and DISA) – Committed to Sensor. ML and SWE as direction • OGC OWS 5. 1 Georeferenceable Imagery (NGA/NASA) – will be demonstrating use of Sensor. ML within JPEG 2000 and JPIP for support of georeferencing of streaming imagery • Oak Ridge National Labs Sensor. Net – funded project will be adding Sensor. ML support in Sensor. Net nodes for threat monitoring (including georeferenced streaming video) • Empire Challenge 2008 (NGA) – planning to test Sensor. ML for sensor observation processing and fusion in 2008 testbed • European Space Agency – developing Sensor. ML profiles for supporting sensor discovery and processing within the European satellite community • Hurricane Missions (NASA MSFC) – working toward using Sensor. ML for geolocation and processing of satellite and airborne sensors during real-time missions • Sensors Anywhere (SAny) – intending to use Sensor. ML/SWE Common within large European sensor project • NASA– funded 30 3 -year projects developing capabilities for Sensor. ML and Sensor Webs; Also recently announce call for SBIR proposals with Sensor. ML and Sensor Web topics identified Mike Botts – January 2008 23
Pulse. Net: Sensor. ML-Enabled Discovery, Data Access, and Tasking Credit: Northrop Grumman Pulse. Net Project Mike Botts – January 2008 24
NASA Projects: Sensor. ML-Enabled On-demand Processing (e. g. georeferencing and product algorithms) AMSR-E SSM/I TMI & MODIS footprints MAS TMI Geolocation of satellite and airborne sensors using Sensor. ML Mike Botts – January 2008 Cloudsat LIS 25
Sensor. ML – Sensor Systems System - Aircraft IR radiation Sensor 1 Scanner Digital Numbers Attitude Sensor 2 IMU Pitch, Roll, Yaw Tuples Location Sensor 3 GPS Lat, Lon, Alt Tuples Mike Botts, Alexandre Robin, Tony Cook - 2005 Mike Botts – January 2008 26
AIRDAS UAV Geolocation Process Chain Demo AIRDAS data stream (consisting of navigation data and 4 -band thermal-IR scan-line data) Mike Botts – January 2008 AIRDAS data stream geolocated using Sensor. ML-defined process chain (software has no a priori knowledge of sensor system) 27
Sensor. ML for Discovery Mike Botts – January 2008 28
Sensor. ML provides metadata suitable for discovery of sensors and processes Find all remote sensor systems measuring in the visible spectral range with ground resolution less than 20 m. Mike Botts – January 2008 29
Discovery Based on Sensor. ML Credit: Northrop Grumman Pulse. Net Project Mike Botts – January 2008 30
Specific Discovery Needs • Unique resource ids used throughout SWE; – sensor centric example: • Find sensors that can do what I need (returns id=“urn: ogc: id: sensor: 123”) • Get me a full description of this sensor urn: ogc: id: sensor: 123 • Now, find a service (SPS) that lets me task this sensor urn: ogc: id: sensor: 123 • Find all services (SOS) where I can get observations from this sensor urn: ogc: id: sensor: 123 • Find all processes that can be applied to this sensor output to generate the information I require • Catalog profiles for each need: – SPS, SOS, SAS services – sensors and processes – observations – terms (either through dictionaries or ontologies) Mike Botts – January 2008 31
Need for Term Definitions used in Sensor. ML and SWE • • • Observable properties / phenomena / deriveable properties (“urn: ogc: def: property: *) – temperature, radiance, species , exceeding. Of. Threshold, earthquake, etc. – rotation angles, spectral curve, histogram, etc. Capabilities, Characteristics, Interfaces, etc. (“urn: ogc: def: property: *”) – Width, height, material composition, etc. – Ground resolution, dynamic range, peak wavelength, etc. – RS-232, USB-2, bit. Size, baud rate, base 64, etc. Sensor and process terms (“urn: ogc: def: property: *”) – IFOV, focal length, slant angle, etc. – Polynomial coefficients, matrix, image, etc. Identifiers and classifiers (“urn: ogc: def: identifier. Type: *; urn: ogc: def: identifier: *” ) – Identifiers – long. Name, short. Name, model number, serial number, wing. ID, mission. ID, etc. – Classifiers – sensor. Type, intended. Application, process. Type, etc. Role types (“urn: ogc: def: role: *”) – Expert, manufacturer, integrator, etc. – Specification document, product. Image, algorithm, etc. Sensor and process events (“urn: ogc: def: classifier: event. Type: *”) – Deployment, decommissioning, calibration, etc. Mike Botts – January 2008 32
Help, Help • We need authoritative bodies with access to subject-matter-experts (SME) to step forward to establish resolvable term dictionaries for sensors, processes, and observations • • • Potential authoritative bodies – IC community – GIG, MASINT WG, Multi-INT Interoperability Lab ? ? – Civilian satellite community – Committee for Earth Observation Satellites (CEOS) – Others - ? ? ? Way forward – Create namespace for terms – Develop web interface for submitting term (Wikipedia perhaps, XML-based? ) • Term • Definition • References • Relationship (? ) – or allow separate ontologies to provide this • Level of authorization – Set up web services for resolving and getting filtered list of terms – Set up authentication process and authentication levels (e. g. submitted, under review, approved, rejected) Accepting Sensor. ML and SWE without creating authorized terms won’t accept interoperability Mike Botts – January 2008 33
Where and how Sensor. ML can be used Mike Botts – January 2008 34
Supports description of Lineage for an Observation Sensor. ML Observation Within an Observation, Sensor. ML can describe how that Observation came to be using the “procedure” property Mike Botts – January 2008 35
On-demand processing of sensor data Sensor. ML Observation Sensor. ML processes can be executed on-demand to generate Observations from low-level sensor data (without a priori knowledge of sensor system) Mike Botts – January 2008 36
On-demand processing of higher-level products Sensor. ML Observation Sensor. ML processes can be executed ondemand to generate higher-level Observations from low-level Observations (e. g. discoverable georeferencing algorithms or classification algorithms) Mike Botts – January 2008 37
On-demand Geolocation using Sensor. ML AMSR-E SSM/I TMI & MODIS footprints MAS TMI Geolocation of satellite and airborne sensors using Sensor. ML Mike Botts – January 2008 Cloudsat LIS 38
Clients can discover, download, and execute Sensor. ML process chains Sensor. ML-enabled Client (e. g. STT) SLD Sensor. ML Open. GL SOS Stylers For example, Space Time Toolkit is designed around a Sensor. ML front-end a Styler backend that renders graphics to the screen Mike Botts – January 2008 39
Incorporation of Sensor. ML into Space Time Toolkit being retooled to be Sensor. ML process chain executor + stylers Mike Botts – January 2008 40
Space Time Toolkit Sample Applications -2 - Mike Botts – January 2008 41
Sensor. ML-Enabled Web Services Mike Botts – January 2008 42
Sensor. ML can support generation of Observations within a Sensor Observation Service (SOS) SOS Web Service Sensor. ML request Observation For example, Sensor. ML has been used to support on-demand generation of nadir tracks and footprints for satellite and airborne sensors within SOS web services Mike Botts – January 2008 43
Incorporation of Sensor. ML into Web Services Sensor. ML process chains have been used to drive on-demand data within services (e. g. satellite nadir tracks, sensor footprints, coincident search output) Mike Botts – January 2008 44
Sensor. ML can support tasking of sensors within a Sensor Planning Service (SPS) SPS Web Service Sensor. ML request For example, Sensor. ML will be used to support tasking of video cam (pan, tilt, zoom) based on location of target (lat, lon, alt) Mike Botts – January 2008 45
SPS control of Web Cam Mike Botts – January 2008 46
Sensor. ML for Portrayal Mike Botts – January 2008 47
SWE Visualization Clients can render graphics to screen Sensor. ML-enabled Client (e. g. STT) SLD Sensor. ML Open. GL SOS Stylers Mike Botts – January 2008 48
SWE Portrayal Service can “render” to various graphics standards SWE Portrayal Service SLD Sensor. ML KML Collada SOS Stylers Google Earth Client For example, a SWE portrayal service can utilize a Sensor. ML front-end a Styler back-end to generate graphics content (e. g. KML or Collada) Mike Botts – January 2008 49
Sensor. ML to Google Earth (KML – Collada) AMSR-E SSM/I MAS TMI LIS Mike Botts – January 2008 50
Sensor. ML Support Activities Mike Botts – January 2008 51
Current Tool Development and Support for Sensor. ML • Sensor. ML Forum – mail list for Sensor. ML discussion (250+ active members from various backgrounds) http: //mail. opengeospatial. org/mailman/listinfo/sensorml • Open Source Sensor. ML Process Execution Engine – Along with open-source process model library, provides execution environment for Sensor. ML described algorithms • Open Source Sensor. ML editor and process chain development client – on-going development of tools to allow human-friendly editors for Sensor. ML descriptions • Sensor. ML-enabled decision support client – Open source Space Time Toolkit is Sensor. ML-enabled and will be available to discover, access, task, and process sensor observations; use as is or as template for COTS development • Sensor. ML white papers and tutorials – being written and released on an array of Sensor. ML topics Mike Botts – January 2008 – Describing a Simple Sensor System – Creating New Process Models; 52
Sensor. ML Process Editors Currently, Sensor. ML documents are edited in XML (left), but will soon be edited using human friendly view (below) Currently, we diagram the process (right top) and then type the XML version; soon the XML will be generated from the diagram itself (right bottom) Mike Botts – January 2008 53
Simple Sensor. ML Forms for the Mass Market User fills out simple form with manufacturer name and model number, as well as other info. Then detailed Sensor. ML generated. Mike Botts – January 2008 54
How can you help? • Investigate Sensor. ML for you needs – Can it meet some of your needs? – What would be your main initial focus (discovery, lineage, processing, display)? • Sensor. ML Forum – participate in the forum; ask/answer questions • Term Dictionaries – help develop online dictionary of sensor terms used in your community • Sensor and process profile – develop Sensor. ML profiles for your sensors and processes • • Test and demonstrate – Test Sensor. ML applications for your environment – Demonstrate successes – Feedback failures, suggestions, and additional needs Participate in the OGC process Mike Botts – January 2008 – Join and attend Technical Committee meeting (particularly SWE WG) – Sponsor Interoperability Projects in OGC 55
Relevant Links Open Geospatial Consortium Standard Documents http: //www. opengeospatial. org/standards OGC Approved Schema http: //schemas. opengis. net/ Sensor Web Enablement Working Group http: //www. opengeospatial. org/projects/groups/sensorweb Sensor. ML information http: //vast. uah. edu/Sensor. ML Public Forum http: //mail. opengeospatial. org/mailman/listinfo/sensorml Mike Botts – January 2008 56
- Slides: 56