Nils J. Nilsson, a computer scientist who helped develop the first general purpose robot and was a co-inventor of algorithms that made it possible for the machine to move about efficiently and perform simple tasks, died on Sunday at his home in Medford, Ore. He was 86.
His death was confirmed by his wife, Grace Abbott.
Dr. Nilsson was a member of a small group of computer scientists and electrical engineers at the Stanford Research Institute (now known as SRI International) who pioneered technologies that have proliferated in modern life, whether in navigation software used in more than a billion smartphones or in such speech-control systems as Siri.
The researchers had been recruited by Charles Rosen, a physicist at the institute, who had raised Pentagon funding in 1966 to design a robot that would be used as a platform for doing research in artificial intelligence.
Although the project was intended to create a general purpose mobile “automaton” and be a test bed for A.I. programs, Mr. Rosen had secured the funding by selling the idea to the Pentagon that the machine would be a mobile sentry for a military base.
At one Pentagon meeting he was asked if this automaton could carry a gun. “How many do you need?” he answered. “I think it should easily be able to handle two or three.”
Image
One of several books by Dr. Nilsson, “The Quest for Artificial Intelligence” was published in 2010 by Cambridge University Press.
The researchers puzzled about what to name their robot, then decided that because it “shook like hell” when it moved, they would just call it Shakey. In 1970, Life Magazine, overstating its abilities, called the machine “the first electronic person” and suggested that true “thinking” machines would arrive in the near future.
Dr. Nilsson, who had specialized in radar, joined the Stanford institute in 1961. Another member of the group, Dr. Peter Hart, recalled in an interview that in recruiting Dr. Nilsson, Mr. Rosen had poked his finger at Dr. Nilsson’s chest and said: “Radar? That’s like doing research on light bulbs! You have to come help us design these learning machines.”
An early focus of Dr. Nilsson’s work involved neural networks, a new technology at the time that had been pioneered by Frank Rosenblatt at Cornell University. That technology would fall out of fashion in the 1970s, then re-emerge this decade after the cost of computing and gathering large data sets fell dramatically.
With the addition of vast amounts of data, neural networks began to rival human qualities in speech understanding and vision.
In 1965, Dr. Nilsson published one of the first books in the field of neural networks, “Learning Machines: Foundations of Trainable Pattern-Classifying Systems.” The approach broke with the dominant direction of artificial intelligence at the time.
Edward Feigenbaum, an early member of the artificial intelligence research community, called the book “revelatory.”
One challenge for the Stanford researchers was to figure out how a robot might navigate in an environment full of obstacles. Dr. Nilsson collaborated with Bertram Raphael and Dr. Hart to create what became known as the A* (pronounced “A Star”) algorithm, which allowed Shakey to find the shortest path between two points in a room strewn with obstacles.
Dr. Hart recalled walking down a hallway and encountering his two fellow researchers deep in discussion about how to calculate the most efficient route for a robot. He went home that evening and spent hours thinking about finding a mathematical proof that would show that a given path was the shortest one possible. He returned the next day and began working with his colleagues to come up the A* algorithm.
Dr. Nilsson worked with another researcher, Richard Fikes, to develop an algorithm to do higher-level planning, or reasoning, known as the Stanford Research Institute Problem Solver, or Strips. The program was designed to enable Shakey to perform simple tasks like finding and moving blocks; this required the machine to reason through a problem in an abstract way.
After the Shakey project wound down, Dr. Nilsson gained funding for another project at SRI International, known as the “Computer-based Consultant,” which focused on natural language understanding. It was a predecessor to Siri, which was also launched at SRI and spun off as an independent company in 2007 before it was acquired by Apple in 2010.
Dr. Nilsson was named chairman of the Stanford computer science department in 1985. He was also the author of “The Quest for Artificial Intelligence: A History of Ideas and Achievements” (2010), among other books.
Nils John Nilsson was born on Feb. 6, 1933, in Saginaw, Mich., to Walter and Pauline (Glerum) Nilsson. When he was 11, the family moved to California, where his father was a salesman for an industrial equipment distributor. His mother was a homemaker.
Dr. Nilsson studied at Stanford as an undergraduate and in 1958 received his Ph.D. there in electrical engineering. His dissertation was in information theory, exploring the problem of both detecting and jamming radar.
He then joined the Air Force and served three years, stationed at the Rome Air Development Center in Rome, N.Y., a research laboratory, before joining the Stanford Research Institute.
In 1958 he married Karen Braucht, who died in 1991. Along with his wife, Ms. Abbott, his survivors include two children from his first marriage, Lars Nilsson and Kristin Nilsson Farley; four stepsons; four grandchildren; and eight step-grandchildren.
Unlike many engineers and computer scientists in Silicon Valley, Dr. Nilsson shied away from the start-up frenzy that has been emblematic of the region.
“He was a researcher and academically inclined,” Dr. Hart said. “He didn’t have any interest in commerce or, as he would call it, ‘industry.’ That was great for other people, but he was not interested.”
