Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
SAF Archives  search ask the scientists in the classroom cool science
scientists from previous shows
cool careers in science
ask the scientists
PREVIOUS SCIENTISTS

Photo of Maja Mataric Maja Mataric as seen on Natural Born Robots: Go, Team!

Click on Maja's photo to read a brief bio.



q I saw you and your robots on Frontiers, and I would like to make a robot like yours (but not ducklings, I would like to make a dog). Do you have any suggestions for websites, catalogs or books that could help me? (Similar question sent by many viewers)

A I teach an introductory robotics course at USC, and have put together a very informative Web page for it, which includes a great deal of information on all aspects of robotics. I suggest you look there: http://www-scf.usc.edu/~csci445/ (Look under "Course Resources") Note that Sony already has a very popular (in fact, sold-out) version of a robot dog, and it took many years of work by many researchers to make it what it is, and it is not as advanced as most of us would like it to be. So it is a very, very hard problem.


q I'm curious what you're going to do next with this robot design. For example, what new capabilities are you thinking of adding? (Question sent by several viewers)

A Since my work is on robot minds and behavior, not design, my future goals are all about how to get groups of robots to interact in social, productive, and adaptive ways. I have ongoing research in various directions to cover those goals, including having robots imitate each other as well as humans, in order to learn new skills and generally interact more naturally; also I have a project on having a human teach a robot by having it observe its behavior as it cooperates with the robot; finally I have several projects on getting groups of robots to automatically adapt to changing environments, tasks, and interactions as they go about their work.


q You say the goal of your research is to get robots to interact in social, productive and adaptive ways. But what, if anything, do you predict robots won't be able to do, even with the most modern technology available to you and your crew of scientists? (Question sent by Anoush)

A I doubt that robots will ever act in ways that are uniquely human, including human irrational and emotional behavior. That is not to say that we won't be able, some day, to build emotional (and even irrational, if we want) robots; I suspect we will. But they will never have those properties in the same way that humans do, because all these things are deeply related to being embodied in a particular body, which includes everything from neurons and hormones to flesh and bones. However, robots will have their own complex brains and mechanisms, so it will be fascinating to see what kinds of properties they will display as they become increasingly more complex.


q Will robots ever be intelligent enough to hold a real conversation with people, or think on their own like we do? (Question sent by Kathryn, Holliston Middle School, Massachusetts)

A Yes, I believe that robots will be able to converse with humans. They will not have the same topics to discuss as people do, nor the same world view or opinions, but they'll be able to interact, exchange information, and even "converse". As for thinking on their own, robots already do that now. Any autonomous system thinks on its own. It's just that currently our systems don't yet think about or do very complicated things, but that is not the same as not thinking at all.




 

Scientific American Frontiers
Fall 1990 to Spring 2000
Sponsored by GTE Corporation,
now a part of Verizon Communications Inc.