Note to users. If you're seeing this message, it means that your browser cannot find this page's style/presentation instructions -- or possibly that you are using a browser that does not support current Web standards. Find out more about why this message is appearing, and what you can do to make your experience of our site the best it can be.
STEM Talent Event

Site Tools

  • AAAS
  • Subscribe
  • Feedback

Site Search

Search Advanced

Science 28 May 1999:
Vol. 284. no. 5419, p. 1457
DOI: 10.1126/science.284.5419.1457d

Random Samples

It sits, it barks, it wags its tail. It can chase a ball, but it never leaves messes (although it does pretend to piddle). It's what Sony calls the world's first entertainment robot capable of autonomous behavior.

About the size of a toy poodle, Aibo is stuffed with sensors and actuators. Switch it on and the battery-powered animal will scratch, yawn, explore its environment, or play with a bright pink ball. It will also sit or lie down in response to patterns of tones from a handheld remote controller. Aibo can't recognize voice commands but he can learn, thanks to artificial intelligence software that enables his behavior to evolve over time. Slap the top of its head after it piddles, and it will do so less often. Pat it after it sits and wags its tail, and it will be progressively inclined to sit and wag. The electronic dog maintains its balance using feedback from an accelerometer, a gyroscope, and pressure sensors in its paws. Legs are assemblies of gears that allow Aibo to mimic a real dog's crouching, walking, and scratching with amazing fidelity.

It won't be long before robots become "bigger than the computer industry," predicts Toshitada Doi, head of Sony's Digital Creatures Laboratory. Sony says the Aibo Web site is already being accessed by about 74,000 people a day. The company is preparing to unleash 5000 Aibos for $2500 each, available over the Internet as of 1 June.





ADVERTISEMENT
Click Me!

ADVERTISEMENT
Click Me!

To Advertise     Find Products


Science. ISSN 0036-8075 (print), 1095-9203 (online)