ASU plays role in making spacecraft more autonomous


TEMPE, Ariz. NASA's ambitious project to make its spacecraft more autonomous includes software algorithms developed by Arizona State University planetary scientists. NASA's Autonomous Sciencecraft Experiment (ASE) promises to give a spacecraft the capability to process data on-board, detect changes and trigger an automated response, such as taking additional images of an area experiencing change.

This is a dramatic departure from how control over spacecraft is handled today, which typically is done by a team of researchers on Earth. The goal is to make the spacecraft more self-reliant for discerning interesting sites for scientists on Earth. That way it can focus on the areas of interest, and immediately respond to dynamic events, eliminating time delays that arise from transmitting data back to Earth, processing and analyzing the data, and only then sending new instructions to the spacecraft.

"The Autonomous Sciencecraft Experiment is a big step in the right direction to creating fully autonomous spacecraft," said Steve Chien, the JPL senior technologist for software on the ASE project. "When operating in the outer reaches of the solar system, it can react quickly to important science events and quickly send the processed data back to Earth."

ASE consists of three computer programs. The first is a planner that organizes the spacecraft activity, working from a list of science goals. The second part is a specialized code that controls the spacecraft and instruments. The third part is made up of programs that search the data for scientifically interesting features.

ASU's contribution is the development of algorithms to detect changes in ice on Earth, such as the freeze and thaw of lakes, snowfall on land, and the formation and retreat of polar sea ice.

"We want to demonstrate we can do this," said Thomas Doggett, a graduate student and member of ASU's Planetary Geology Group. Doggett and ASU professor Ron Greeley are working with NASA scientists on ASE. "The first step of this process is to develop algorithms that can discern differences between snow, water, ice and land."

The ASU algorithms were uploaded to Earth Observing-1, a satellite operated by NASA's Goddard Space Flight Center, which carries Hyperion, a hyper-spectral visible/infrared imager. In addition to the Arizona State ice algorithm, ASE is studying volcanic and flooding activity with algorithms developed by scientists at JPL and the University of Arizona, respectively.

The ASU algorithm had its first complete test on April 29th, when sea ice detected in an image of Barrow Channel near the town of Resolute in the Canadian arctic triggered an observation of Ward Hunt Ice Shelf on the following day. In addition to sea ice in Canada and Alaska, the ice algorithm is being tested on lakes in Wisconsin, Minnesota and Tibet, as well as sea ice in Antarctica.

"Autonomous detection of ice change is important for spacecraft monitoring a wide range of bodies in the solar system," Doggett said. "For example, evidence of ice has been found in the poles of Mercury and Earth's moon, sheltered from the boiling daytime temperatures by the permanent shadow of crater rims.

"The polar caps of Mars advance and retreat with the seasons much like those on Earth, except they are composed of carbon dioxide ice as well as water ice," he added. "Geysers of nitrogen ice erupt on Neptune's moon Triton, and the geologic study of the outer solar system has shown evidence for other examples of cryo-volcanism, where molten ice acts on icy worlds much as lava does on rocky worlds like Earth."

Another question that could be addressed is whether under the crack-filled icy surface of Europa (a moon of Jupiter) is an ocean warmed by tidal heating from Jupiter, perhaps even supporting life.

"If there is still an ocean today, the answer could come from cracks forming on the surface under the ongoing tidal forces exerted by Jupiter, briefly exposing the material below," Doggett said. "If future spacecraft, like the proposed Jupiter Icy Moons Orbiter, or a Mars mission looking at the polar caps, are equipped with ASE-like functions, they could provide a dynamic view of these phenomena."

ASE is directed by the Artificial Intelligence Group at the Jet Propulsion Laboratory (JPL), Pasadena, Calif. JPL is managed for NASA by the California Institute of Technology.

Source: Eurekalert & others

Last reviewed: By John M. Grohol, Psy.D. on 21 Feb 2009
    Published on All rights reserved.