National Aeronautics and Space Administration
Small Business Innovation Research 2001 Program Solicitation

TOPIC E3 Advanced Information Systems Technology

[ back to Solicitation ] [ back to Chapter 8.1.1 ] [ back to Chapter 8.1 ] [ back to Chapter 8 ] [ back to table of contents ]

E3.01 Knowledge Discovery and Data Fusion
E3.02 Automation and Planning
E3.03 High Performance Computing and Networking
E3.04 Geospatial Data Analysis Processing and Visualization Technologies
E3.05 Data Management and Visualization

The Earth Science Enterprise (ESE) acquires, processes and delivers very large (gigabyte to terabyte) volumes of remote sensing and related data to public and government entities that apply this information to understand and solve problems in Earth science. Information technology is currently employed throughout ESE's space and ground systems, and the Advanced Information System Technology theme is soliciting technologies that apply to the end-to-end system functions. The information system functions found in ESE include but are not limited to data acquisition, data transmission, data processing, data management and storage, data distribution, data/metadata/document search, browse and assess, data subsetting, knowledge discovery, spatial-temporal analysis, and visualization. The ESE is interested in advanced information technology that can improve any of these functions in isolation or in combination or is able to support alternative architectures that better address the scientific requirements.

E3.01 Knowledge Discovery and Data Fusion
Lead Center: JPL
Participating Center(s): None

NASA's Earth Science Enterprise collects terabyte-scale datasets routinely during its missions and charges the scientific community with extracting usable and scientifically relevant information from them. These data sets may be images, multispectral images, time series, or field and particle event lists. They may also be engineering time series about spacecraft health collected from on-board sensors. Emphasis has recently been placed on handling and analyzing in situ data from networks or sensorwebs. In addition to the ongoing challenges entailed by handling, analyzing and mining very large data sets, NASA now needs a new framework for performing science data evaluation on-board spacecraft and from in situ sensor networks. New on-board or in situ science capabilities will enable mission activities to be directed by scientists without the assistance of a ground sequencing team and the constraints of communications links. The science capabilities will be adaptive in nature and must be efficient in transmission of the usable key data.

This subtopic enlists help in developing a new generation of tools and algorithms for effective acquisition and analysis of data and image sets appropriate for ground or on-board/in situ use. Of special interest are: (1) the ability to deal quantitatively with uncertainty present in data, perhaps in a statistical framework; (2) development of flexible models through which observables are linked to quantities of scientific or engineering interest; (3) harnessing database technology for organizing the observed data, models, and inferred knowledge, perhaps in on-board or in situ archives; (4) fusion of multiple datasets for enhanced scientific return; and (5) system concepts for handling interactions between on-board science analysis and event detection capabilities and other functions of an autonomous spacecraft or sensor web. One or more of these areas should be addressed by every proposal. Specific technologies of interest would address:

• Automated classification of data
• Supervised and unsupervised learning methods
• Knowledge discovery techniques
• Image analysis and segmentation
• Statistical pattern recognition
• Time series feature extraction and analysis
• Trainable object recognition
• Automatic image registration and change detection
• Visualization and rendering techniques
• Spatial-temporal data mining
• Intelligent, goal-directed data acquisition and/or compression
• Science data analysis algorithms designed for scalable computing
• System concepts for on-board science
• Adaptive data acquisition techniques

E3.02 Automation and Planning
Lead Center: ARC
Participating Center(s): None

Focus is on technologies that make a spacecraft or system react to uncertainties in a robust fashion while achieving a set of high-level goals or tasks. Technology innovations in automation and autonomous systems are required to support the high level command collection and effective techniques for processing large volumes of data into useful information. Intelligent data discovery and searching over heterogeneous data in distributed data stores. Collaboration between Earth scientists and computer scientists is encouraged for these proposals to demonstrate useful results. Areas of interest in technological innovation include:

• Autonomous agents: Intelligent autonomous mobile search agents to support science applications involving data available on the Internet
• Autonomous data collection: Automatic dynamic reconfiguration of UAV or space on-board data gathering instruments to make effective use of observing conditions, baseline image data priority scheme, history of observations and limited on-board resources
• Planning and scheduling
• System health and maintenance (space and ground based)
• Distributed decision making (multiple agents, autonomous systems)
• Automated software testing
• Legacy code maintenance and conversion
• Automatic software generation (i.e., processing algorithms)
• Software tools for parallelization; tools for production planning
• Control of FPGA to provide real-time products using hyper-spectral instrument data from airborne platforms
• Verification and validation of automated systems

E3.03 High Performance Computing and Networking
Lead Center: ARC
Participating Center(s): GSFC

This subtopic focuses on innovations in efficient and effective techniques and technologies for processing large volumes of data into useful information. Areas of interest include:

• High performance processing
• Computing: Distributed computing, Reconfigurable computing, Parallel/cluster computing, Embedded computing, Optical computing
• Future computing and storage device technologies: quantum computing, atomic chain electronics, molecular computing, nano and quantum device technologies, and carbon nanotube based electronic devices and proposed architectures
• Innovative node connection networks
• High performance/pervasive networks
• Techniques to enhance performance of wide-area networks supporting highly distributed data production, archive, and access functions
• High-speed processing architectures/systems; applications of distributed computing environments, especially "pervasive computing"
• Efficient methods/algorithms/systems for warehousing scientific multispectral and hyperspectral data and/or instrument data for automatic and user-directed mining/monitoring of meaningful trends, parameters, fluctuations, etc. to maximize scientific value of TB-sized data sets
• Facilitating portability across architectures
• Advanced Storage and archival techniques (e.g., 3D holographic memory, holographic storage)
• Load balancing techniques
• Standards to simplify data providers' activities while facilitating data usage by a large user community
• Server side technologies supporting highly responsive user-centric access (e.g., handheld PDAs to large date centers)
• Software development environments and methodologies
• Work scheduling as applied to distributed computer systems
  

[back to top]

E3.04 Geospatial Data Analysis Processing and Visualization Technologies
Lead Center: SSC
Participating Center(s): None

Proposals are sought for the development of advanced technologies to enhance human and machine interaction in support of scientific, commercial and educational application of remote sensing data. An emphasis is on distributed and/or mobile teams in validation and verification exercises and for the commercialization of remote sensing data. Focus areas are to provide tools for interpretation, visualization or analysis of remotely sensed data and to provide qualitative and quantitative analysis tools and techniques for performance analysis of remotely sensed data. Applications can support the commercial remote sensing industry and enhance the commercial or educational application of Earth science data. Areas of specific interest include:

• Unique, innovative data reduction and rapid analysis methodologies and algorithms, particularly for hyperspec-tral data sets
• Innovative techniques for validation of imaging systems (i.e. thermal and LIDAR imaging systems)
• Software tools for mobile computing and efficient data collection and/or presentation
• Innovative approaches for incorporation of GPS data into in situ data collection operations with dynamic links to spatial databases including environmental models
• Innovative techniques to automate quality assurance processes for science data products
• Distribution and sharing of fused science data sets to correlate similar data sets from diverse spacecraft and aerial vehicles and provide unique, commercially useful information products
• Data merge and fusion software for efficient production and real-time delivery of commercial digital products to teams and remote users
• Tools for enabling distributed scientific collaboration
• Software to automate the rapid processing and distribution of sub-setting and presenting RS data over a network
• Software to develop commercial products from digital topography and vegetation canopy data obtained from airborne and space-based active optical sensors
• Innovative approaches that contribute to the understanding of data through the display and visualization of some or all of the above data types including providing the linkages and user interface between the cartographic model and attribute databases
• Visualization of multivariate geospatial data including remotely sensed data from the following:
• airborne and satellite platforms, vector data from public and private archives;
• cartographic databases from public and private sources;
• continuous surface data held as a raster data model; and
• 3-D data held in a true 3-D raster model.
  

[back to top]

E3.05 Data Management and Visualization
Lead Center: GSFC
Participating Center(s): None

This subtopic focuses on innovative approaches to locating, summarizing and presenting large collections of Earth science data in a highly distributed and networked environment. Collaboration between Earth scientists and computer scientists is encouraged for the proposals to demonstrate useful results. Specific examples of topics of interest are:

• Design and implementation of a virtual reality CAVE for scientific data visualization Ideas can include: 3-D virtual reality environments that will let users 'fly' through the data space; precomputed data fly-throughs that let users search within the fly-through space (i.e. fast forward, reverse, slow motion) to locate specific areas of interest; incorporation of commodity data compression techniques (such as HDTV/MPEG) for reduced storage and transmission requirements; progressive compression and caching techniques that optimize resolution and performance when zooming in for additional detail; techniques for georectification, data overlays, data reduction, and data encoding that work across a distributed environment of widely differing data types and formats; development of integrated object oriented storage and compression techniques that are integrated into search algorithms; novel 3-D presentation techniques that minimize or eliminate the need for special user devices such as goggles or helmets; techniques for high bandwidth collaboration with other users in a distributed environment; development of techniques that invoke integrated visual and auditory presentation cues; data viewing and real-time data browse, including fast, general purpose rendering tools for scientific applications; viewing of multivariate geospatial data including remotely sensed data.
• Support a collaborative environment with tools that facilitate outreach and problem solving. This environment stimulates government & business partnerships with emphases on detailed data analysis conversion capabilities to translate science & technology data into information used by specialty communities for making decisions.
• Tools for enabling distributed scientific collaboration.
• Technologies supporting management, storage, search and retrieval of very large, distributed, geo-spatial earth science data volumes: Tools to facilitate automatic data product legacy, quality assurance and metadata updates. Object relational technologies specific for Earth sciences. Meta-data discovery to facilities the automated use of data from different sources. Automatic metric collection and analysis for data use and data ordering. Smart Objects Dumb Archives (SODA) and storage, archival and retrieval standards applicable to ESE mission requirements.
  

[Back to top]

[back to top] [ back to Solicitation ] [ back to Chapter 8.1.1 ] [ back to Chapter 8.1 ] [ back to Chapter 8 ] [ back to table of contents ]