'Thinking' robot's job will be to do the dirty work for man

A new "graduate" is expected to replace some of the shrinking number of traditional 18- to 22-year-old job-seekers. The new graduate's opportunities are unlimited.

This graduate routinely takes jobs too dangerous or tedious for other job-seekers. Whether working on an assembly line or in a "white collar" job, the graduate always uses good judgement in "thinking" through the task at hand.

This new graduate is a robot, from the school of robotics.

What careers will these graduate robots pursue? According to researchers at the Robotics Institute at Carnegie-Mellon University in Pittsburgh, a new school devoted solely to the study and application of robot technology, the most likely "careers" for robots will be:

* Hazardous-environment work -- where conditions are dangerous or impossible for human beings -- such as alongside for human beings -- such as longside blast furnaces, inside nuclear power and chemical plants, or working in undersea exploration.

* Defense-related tasks -- as mobile scouts on one-way "no return" assignments, in underwater surveillance or demolition assignments; on satellite spy-intercept missions.

* Space work -- making repairs in the hostile environment of space, exploring and economically exploiting other planets.

* "Blue collar" work -- spot welding on auto assembly lines, using perception and judgment in assembling parts, handdling materials, and checking quality control.(Industrial robots are already at work in the United States, notably on new auto assembly lines. But Japan, experiencing a shrinking number of young workers, possesses more than half the world's industrial robots and is considered the world leader in developing industrial robots that build other robots.)

* Medical fields.

To help fund its robot research, now at $3 million annually, Carnegie-Mellon relies on a cooperative arrangement between the Robotics Institute and industry. Joint use of any robotics research is made between the parties involved. All patent rights that might arise from its interdisciplinary approach go to a principal industrial sponsor. The university retains all publishing rights from any research.

The institute hopes to have a single sponsor for each of the above areas of research. Westinghouse is the principal sponsor in the blue-collar area ($1 million a year for five years).

Daniel Berg, dean of the Mellon college of Science, says "breakthrough are most likely to occur when the expertise of several disciplines is focused on achieving a definite goal -- like getting men to the moon. Our approach to robotics research necessarily includes not just an interdisciplinary stdy of public-policy issues and business economics but a multi-disciplinary one."

From its inception, the Robotics Institute at Carnegie-Mellon has been run by a three-member board to assure an interdisciplinary approach: Dr. Berg; engineering school dean Angel G. Jordan; and Raj Reddy, director of the Robotics Institute and a professor of computer science, who developed a 1,000-word speech-recognition system that is 98 percent accurate with the words used in context.

In addition to the 12,500 square feet the university has given the Robotics Institute, a new facility to open next month will provide an additional 60,000 square feet for expansion.

Purdue University in Lafayette, Ind., already has a "student" robot attending classes. It is a T is the central element in the interdisciplinary development of engineering fundamentals that will lead to the installation of similar equipment in operating manufacturing systems.

A "blue collar" robot, the T is capable of lifting up to 255 pounds and performs tedious as well as hazardous tasks. At a Lockheed-Georgia aircraft plant it presents web assemblies for airplanes floor structures -- two at time -- to a machine for drilling and riveting.

Like many of its future "classmates," the T mimics human actions. But any physical resemblance ends with the fingertips on its giant arm. Designed soley for the workplace, it doesn't need legs because it doesn't go home at 5 o'clock.

In contrast, "white collar" robots, such as those the computer science department at Stanford University in California is working to develop, will be intelligent "assistants" in scientific and medical offices.

We want to hear, did we miss an angle we should have covered? Should we come back to this topic? Or just give us a rating for this story. We want to hear from you.