NMSU branding

New Mexico State University

New Mexico State University

News Center




New Mexico State University professor to research robot development

A New Mexico State University professor plans to spend a year-long NASA fellowship researching the development of robots that may someday accompany -- or precede -- humans to Mars and beyond.



Nadipuram Prasad, a professor in New Mexico State University's Klipsch School of Electrical and Computer Engineering, will spend a year at the Jet Propulsion Laboratory's Center for Space Microelectronics Technology researching the artificial intelligenc

Beginning in August, Nadipuram Prasad, a professor in the university's Klipsch School of Electrical and Computer Engineering, will spend a year at the Jet Propulsion Laboratory's Center for Space Microelectronics Technology in Pasadena, Calif., under NASA's Administrator's Fellowship program. During his one-year fellowship, Prasad will research new technologies in artificial intelligence for sensing, imaging and storing information that may someday lead to what scientists call "autonomous systems" -- self-controlling robots, he said.

NASA scientists are working to develop robots -- improved versions of the Martian landers -- that will be able to navigate the harsh landscape of Mars and other planets with human-like agility while collecting samples, performing tests, then transmitting the data back to Earth, Prasad said.

Teams of such robots could prospect for minerals on other planets, preparing the way for more elaborate robotic "structures" to carry out mining operations. For human astronauts, self-controlled robots could provide such basic services as construction labor, cooking and laundry, even paramedical services, he said.

But to do such things, the robots must be able to mimic human perceptions of vision, sound, smell, taste and touch. They must also be able to "think," "reason," and communicate with each other and with humans, he said.

Prasad, who specializes in areas of computer controlled systems called systems theory and systems science, said in order to install such capacities scientists must look at many areas.

"An intelligent system is one that has the ability to reason and make decisions in the presence of uncertainty. How does the machine collect and combine data from various sensors? How does it interpret or make partial evaluations based on partial data -- something we humans do all the time? To incorporate these capacities, we have to use a combination of technologies in artificial intelligence: 'fuzzy logic' to come up with decisions based on partial information, 'neural networks' which attempt to mimic human memory and recognition and 'genetic algorithms' which sort and prioritize information, so that, given a set of decisions, the machine can decide which is best," he said.

Computers that combine these technologies are used now by security agencies and by insurance companies that use them to detect fraud. But, Prasad said, they are not mobile, as space-traveling robots must be, and scientists are still studying how their components will react to extremes of temperature, such as those on Mars, where surface temperatures vary from approximately minus 220 degrees Fahrenheit to approximately 80 degrees Fahrenheit.

He added that if rugged, reliable autonomous systems can be developed, they would have a number of applications on Earth as well as in space. He and John J. Ellington, a New Mexico State University professor of entomology, plant pathology and weed science, have been researching the design of a robot that could go into cotton fields, take samples of insects, identify them and recommend strategies for controlling the bad ones, he said.

"Joe and I have already addressed the problems of how the machine would identify the insects and how it would store and retrieve information on them. We have not yet solved the problem of how to take samples in the field and have the machine sort between what you want it to identify and extraneous matter like twigs, dust or microbes," Prasad said.

NASA's Administrator's Fellowship program is intended to give NASA employees the chance to teach and conduct research in minority-serving colleges and universities and to expose faculty at the institutions to NASA research and encourage them to establish NASA-related research programs at their institutions.

Prasad is the director of the Rio Grande Institute of Soft Computing (RioSoft), a consortium of four New Mexico universities -- New Mexico State, New Mexico Highlands, New Mexico Tech and the University of New Mexico -- and the University of Texas at El Paso, which seeks to develop artificial intelligence technologies for use in government and industry. Jeffrey T. Drake, a NASA engineer and Prasad's former student, is currently a NASA fellow at New Mexico State and is involved in setting up RioSoft, Prasad said.