skip to main content
Skip header Section
Designing sociable robotsMay 2002
Publisher:
  • MIT Press
  • 55 Hayward St.
  • Cambridge
  • MA
  • United States
ISBN:978-0-262-02510-2
Published:01 May 2002
Pages:
263
Skip Bibliometrics Section
Bibliometrics
Skip Abstract Section
Abstract

Cynthia Breazeal here presents her vision of the sociable robot of the future, a synthetic creature and not merely a sophisticated tool. A sociable robot will be able to understand us, to communicate and interact with us, to learn from us and grow with us. It will be socially intelligent in a humanlike way. Eventually sociable robots will assist us in our daily lives, as collaborators and companions. Because the most successful sociable robots will share our social characteristics, the effort to make sociable robots is also a means for exploring human social intelligence and even what it means to be human.

Recommendations

Joe L. Podolsky

We all know about human-computer interfaces (HCI). We complain about them every day. Even the best of them seem cold, efficient. When they are bad, they are laughable, as well as horrid. Robots come in two types: those that look like machines, and those that are anthropomorphic. The ones that look like machines, we treat like machines, and interact with them through HCI. But what about those that look like us, or like pets__?__ We expect far more from these. We expect them to have emotions. We expect them, like Data in Star trek: the next generation , indeed, like Pinocchio, to strive to be human. Breazeal’s Pinocchio is called Kismet, and she has built it at MIT’s Artificial Intelligence Lab. As she says, “Kismet is special and unique. Not only because of what it can do, but also because of how it makes you feel. Kismet connects to people on a physical level, on a social level, and on an emotional level.” Breazeal includes a CD-ROM in the book, so that readers can see and hear Kismet, and decide for themselves how well Breazeal has engineered emotions into her creation. Some of the basic work in this area was done at Stanford University in the mid-nineties [1]. That work showed how people inferred emotions in televisions and personal computers, objects that made no pretense at being sentient. Breazeal draws on this, and on other research, to make us care about Kismet, and to interact with it as though it were one of us. Readers who are interested in the concept of sociable robots will be happy to focus mainly on the first and last chapters of the book. Breazeal starts with “The Vision of Sociable Robots,” describing what a robot has to have to be sociable: obvious things like “life-like quality” and “human-aware,” and some non-obvious traits like “socially situated learning.” In the last chapter, Breazeal presents what she calls the “grand challenge problems,” on which she will presumably spend the next phase of her career. The list here is curiously human, almost what a mother would wish for a young child: to develop personality, to be able to engage in discourse, to have a sense of self that comes from memory of its experiences, and to recognize people it has met before. Breazeal asks, “What would be required to build a robot that could be a genuine friend__?__” The bulk of the book is about psychology and engineering. The first few chapters describe what psychologists have learned about how people interact, and the body language and cues we all instinctively use to communicate socially. Since Kismet is purposely built to resemble a young child, Breazeal devotes an entire chapter to developmental psychology and infant-caregiver interactions. Later in the book, Breazeal describes how she has actually designed and built Kismet to invoke the “body language” that people expect from other people, techniques needed to create the illusion of being human. She describes Kismet’s subsystems: visual, auditory, and expressive motor. The expressive motor system controls Kismet’s lips, eyes, and ears; there are 14 actuators in its face that enable it to make eight expressions, including anger, sadness, happiness, and surprise. Examples of all these expressions, along with the sound that supports them, are on the CD-ROM. Breazeal gives reasonable detail about both the hardware and software she developed to bring Kismet to life, and presents data from many experiments that she and her colleagues performed. One of the main characters in Isaac Asimov’s robot books is Susan Calvin, who became the “robopsychologist” of the (still fictional) United States Robots Corporation. According to her biography in I, robot [2], Dr. Calvin was born in 1982, only slightly younger than Cynthia Breazeal. Breazeal’s career might well be laid out for her in the pages of Asimov’s books. This book is useful because it focuses on progress driven by scientific investigation, and on results that come from imaginative engineering experiments. But technology is effervescent, driven as much by marketing hype as by technical progress. This book has longevity not because of its science and engineering, not even because Kismet might be the latest killer toy to show up at Comdex. Breazeal, like the fictional Susan Calvin, has laid the foundation for what sociable robots have to be. In doing so, as did Asimov, and as did Gene Roddenberry with his android Data, she raises the questions about the essence of humanity. It’s not just a question of functions and capabilities. We individual humans are uniquely capable, depending on our birth nature and childhood nurture, on the events that shape our lives, and on our inevitable aging. But through all that, we are human. The meaning of this is a subject that has long been studied by theologians, philosophers, and novelists. And now, perhaps appropriately for the 21st century, we are on the threshold of breakthroughs, both in understanding and application, led there by Cynthia Breazeal as the first robopsychologist. Online Computing Reviews Service

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.