Collaborative Telerobotics

Collaborative Telerobotics
We define a ``collaborative telerobot'' as a telerobot simultaneously controlled by many participants, where input from each participant is combined to generate a single control stream. Collaborative Telerobotics (CT) is an innovative approach to distance learning and telepresence. With CT, participants _collaborate_ rather than compete for access to valuable resources such as historical and scientific sites. A scalable infrastructure for CT, compatible with the Internet, will allow large groups of students or researchers to simultaneously participate in remote experiences. For example, CT can allow groups of disadvantaged students to collaboratively steer a telerobot through a working steelmill in Japan or the Presidential Inauguration, and allow groups of researchers to collaboratively move a telerobot around a newly active volcano or a fresh archaeological site.

We have designed a series of Java-based prototype CT systems, including the Tele-Actor and Sharecam/Co-Opticon. CT raises fundamental new research questions in theory, algorithms, and system implementation. We are developing a science base for a scalable IT infrastructure for CT that will advance human-to-human and human-to-computer remote communication.

Research in Collaborative Telepresence is sponsored by the National Science Foundation, Intel Corporation, UC Berkeley's Interactive University Project, and UC Berkeley's Center for Information Technology Research in the Interest of Society (CITRIS).

Prototype CT Systems:

Publications:

    Collaborative Frame Selection: Exact and Distributed Algorithms for a Networked Robotic Camera with Discrete Zoom Levels. Dezhen Song, A. Frank van der Stappen, and Ken Goldberg. (In Preparation). [0.8MB .pdf]

    ShareCam Part I: Interface, System Architecture, and Implementation of a Collaboratively Controlled Robotic Webcam, Dezhen Song and Ken Goldberg, IEEE/RSJ International Conference on Robots and Systems, October 2003. [0.5MB .pdf]

    ShareCam Part II: Approximate and Distributed Algorithms for a Collaboratively Controlled Robotic Webcam, Dezhen Song, Ken Goldberg, and Anatoly Pashkevich, IEEE/RSJ International Conference on Robots and Systems, October 2003. [0.5MB .pdf]

    Exact and Distributed Algorithms for Collaborative Camera Control,
    Dezhen Song, A. Frank van der Stappen, and Ken Goldberg, Fifth Workshop on the Algorithmic Foundations of Robotics, Nice, France. Dec 2002. [0.5MB .pdf]

    Collaborative Teleoperation using Networked Spatial Dynamic Voting,
    Ken Goldberg, Dezhen Song, and Anthony Levandowski, Proceedings of the IEEE, Special issue on Networked Robots. 91(3), pp. 430-439, March 2003. [2.4MB .pdf]

    Collaborative Online Teleoperation with Spatial Dynamic Voting and a Human ``Tele-Actor''
    K. Goldberg, D. Song, Y. Khor, D. Pescovitz, A. Levandowski, J. Himmelstein, J. Shih, A. Ho, E. Paulos, J. Donath, IEEE International Conference on Robotics and Automation, May 2002. [590K .pdf]

    Collaborative Control of Robot Motion: Robustness to Error
    Ken Goldberg and Billy Chen, IEEE/RSJ International Conference on Robots and Systems, October 2001. [450K .pdf]

    Eigentaste: A Constant Time Collaborative Filtering Algorithm,
    Ken Goldberg, Theresa Roeder, Dhruv Gupta, and Chris Perkins, Information Retrieval Journal,4(2), pp. 133-151. July 2001. [800K .pdf]

Contact:
Prof. Ken Goldberg, IEOR and EECS, UC Berkeley, goldberg@ieor.berkeley.edu
Dez Song, IEOR, UC Berkeley, dzsong@ieor.berkeley.edu