Terms like sensor webs, emNets, and telematics all refer to the marriage of sensors and instruments with network-distributed computing resources. Command, control, communications, and general management of networks and sensing systems are an immature but rapidly growing field. There is much work to be done to evolve and mature sustainable sensor web capabilities for our aerospace applications.
The Global Test Range has developed and actively uses several tools of broad value to the airborne sensor web communities. We also collaborate with customers and like-minded system developers to adopt and/or adapt tools and technologies deemed to be of sufficient maturity and effectiveness for our needs.
Below is a partial list and brief description of the technology and innovations used in the Global Range:
Network Computing Cyberinfrastructure: DataTurbine
In the mid-1990s, researchers at NASA Dryden were focused on network-distributed signal processing. The goal was to assemble the results from potentially many processing steps in different ways via managing the flow of live test data and analyses over shared networks. Under a project funded by NASA's Small Business Innovation Research Program, the result was the patented Ring Buffered Network Bus Technology, the enabling idea within what is now the open source DataTurbine middleware.
In terms of modern computing trends, the DataTurbine enables "cloud computing" architecture for test and measurement applications. Its unique time-based caching approach facilitates applications that need to scroll back and forth through time as many experimental systems require. The Global Test Range Development Laboratory has thus implemented its "sensor web" observational strategy for conducting airborne science missions using DataTurbine.
An airborne science measurement campaign can involve one instrument on one aircraft, or it can involve dozens of instruments, several aircraft, and hundreds of participants. Many times the NASA campaign coordinates with complimentary US and International efforts. In order to maximize the return on investment in these measurement campaigns, disparate data sources must be rapidly integrated into a cohesive system that facilitates situational awareness and interaction amongst geographically dispersed research teams and their instruments. A number of significant integration and operations challenges exist, including heterogeneity of instrumentation and complexity of data stream processing.
The core ground-based system at NASA Dryden Flight Research Center that provides data to the globally distributed customer set is a DataTurbine parent server with a half dozen child servers, each running on separate computers. Two of NASA's aircraft have onboard research teams, and each of those aircraft have their own single DataTurbine server onboard.
Instrument and vehicle status data are acquired at the DataTurbine, typically with http-based polling or udp-based monitoring modules attached to a particular server. For airborne DataTurbine servers, udp and http is also used to ingest data via low bandwidth satellite links from DataTurbine servers on the ground. Weather information such as GOES or MODIS satellite imagery (some of it specialized for specific aircraft) are also acquired and cached in DataTurbine. Intermediate demultiplexing modules morph acquired data as necessary into individually addressable parameters. Additional modules allow the use of third party applications to be used for processing such as parameter scaling. Support of external applications such as Google Earth is provided by modules that generate KML files with near realtime flight track and other pertinent information on demand. External applications including web browsers can be used to query data at any time.
The net effect is cyberinfrastructure that supports real-time interaction amongst instruments and decision makers conducting distributed test and measurement operations involving airborne platforms. Notable productivity improvements have been observed, resulting in higher quality observations and greater overall return on investment in NASA's Airborne Science Program. The successful application of DataTurbine in this prototype environment has demonstrated the value of standing up one or more operational capabilities.
Research Environment for Vehicle Embedded Analysis on Linux (REVEAL)
Emerging from an internal study of the technology readiness of Linux and Java for network-centric vehicular data systems, REVEAL is a configurable embedded system for facilitating integration of instrument payloads with vehicle systems and communication links. REVEAL systems currently serve as onboard data acquisition, processing, and recording systems while also serving as a managed gateway for communication with terrestrial systems. Several classes of REVEAL Physical packages have emerged so far (all of them use the same Linux-based software):
- Altair Class: Award-winning PC/104 design that targets high altitude long endurance unpiloted vehicles. This class of full-featured instrument is roughly the size of a loaf of bread and weights 10-20 lbs. NASA has Altair-Class REVEAL systems operating on several airborne science platforms, including ER-2, DC-8, WB-57F, P-3B Orion and G-III. + read more
- NASDAT: The NASA Airborne Science Data Acquisition and Transmission unit (NASDAT) is a repackaged "production" version of the Altair-Class REVEAL system, designed to replace existing navigational recorders on NASA's airborne science fleet and eventually serving as the operational replacement for Altair-Class REVEAL systems. It is updated to house four EMI-isolated Iridium modems. The NASDAT system was designed jointly with NASA Ames Research Center, and is being developed and maintained by NASA Ames. The Global Hawk platforms are planning to incorporate NASDAT systems into their payload network infrastructure.
- REVEAL system development: A number of other classes of REVEAL systems have been prototyped or are being considered. One design path targets smaller vehicles with reduced requirements and size/weight/power constraints such as sailplanes and Aerosonde-class uninhabited vehicles.
Other embedded computer servers and gateways
In order to support the sensor web capabilities on the larger aircraft, the Global Test Range Development Lab develops or employ embedded or ruggedized systems on the aircraft. So-called "N-Channel" systems are embedded Linux computers that manage Iridium links, like REVEAL but without the data acquisition peripherals. Onboard ruggedized servers host onboard data processing, messaging, and web services so that the onboard research teams can communicate with instruments and researchers that are on the ground or on other aircraft.