earticle

논문검색

Development of an Autonomous Dancing Robot

초록

영어

We have developed an autonomously dancing humanoid, based on the Hitec Robonova platform, that is able to perform sequences of smooth movements in synchrony with music. Our system automatically determines the locations of beats from the acoustic signal of music in real-time, and this information is used to coordinate the motions of the robot, enabling it to perform rudimentary dances. The small (14 inch-tall) RoboNova serves as a prototype development platform for a much larger and more sophisticated humanoid, the KAIST Hubo, which is capable of more fluid and human-like movements. But by first developing and testing systems and algorithms on the smaller robot, we reduce the risk of damaging the larger and more costly Hubo. Regardless of the humanoid platform, however, it is difficult to synthesize motions that appear expressive and artistic. We developed our prototype dancing humanoid in order to further investigate and, hopefully, quantify some parameters of creative movement.

목차

Abstract
 1. Introduction
 2. Prior Work
 3. Algorithm components
  3.1. Beat identifier
  3.2 Robot dancer
  3.3 Gesture generation and control
 4. Beat identifier
 5. Robot platforms
  5.1. Intended platform: Hubo
 6. Gesture generation and control
 7. Experiments and Results
  7.1. Beat identification of multiple songs
  7.2. Gesture cross-fading and cycloid interpolation
  7.3. Improved control environment
  7.4. Tempo-dependent gestures
  7.5. Hubo gesture development and transmission
 8. Conclusions and future work
 9. Acknowledgements
 10. References

저자정보

  • David Grunberg Department of Electrical and Computer Engineering
  • Robert Ellenberg Department of Electrical and Computer Engineering, Department of Mechanical Engineering Drexel University Philadelphia, PA
  • In Hyuek Kim Department of Mechanical Engineering Korea Advanced Institute of Science and Technology (KAIST) Daejeon, South Korea
  • Jun Ho Oh Department of Mechanical Engineering Korea Advanced Institute of Science and Technology (KAIST) Daejeon, South Korea
  • Paul Y. Oh Department of Mechanical Engineering Drexel University Philadelphia, PA
  • Youngmoo E. Kim Department of Electrical and Computer Engineering

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.