• DocumentCode
    2945542
  • Title

    A New Perspective on O(n) Mass-Matrix Inversion for Serial Revolute Manipulators

  • Author

    Lee, Kiju ; Chirikjian, Gregory S.

  • Author_Institution
    Department of Mechanical Engineering Johns Hopkins University Baltimore, Maryland, USA kiju@jhu.edu
  • fYear
    2005
  • fDate
    18-22 April 2005
  • Firstpage
    4722
  • Lastpage
    4726
  • Abstract
    This paper describes a new algorithm for the efficient mass-matrix inversion of serial manipulators. Whereas several well-known O(n) algorithms already exist, our presentation is an alternative and completely different formulation that builds on Fixman’s theorem from the polymer physics literature. The main contributions here are therefore adding a new perspective to the manipulator dynamics literature and providing an alternative to existing algorithms. The essence of this theory is to consider explicitly the band-diagonal structure of the inverted mass matrix of a manipulator with no constraints on link length, offsets or twist angles, and then build in constraints by appropriate partitioning of the inverse of the unconstrained mass matrix. We present the theory of the partitioned mass matrix and inverse of the mass matrix for serial revolute manipulators. The planar N-link manipulator with revolute joints is used to illustrate the procedure. Numerical results verify the O(n) complexity of the algorithm. Exposure of the robotics community to this approach may lead to new ways of thinking about manipulator dynamics and control.
  • Keywords
    Fixman’s theorem; O(n); manipulator dynamics; revolute manipulators; Acceleration; Constraint theory; Heuristic algorithms; Manipulator dynamics; Matrix decomposition; Mechanical engineering; Partitioning algorithms; Robot kinematics; Smoothing methods; Transmission line matrix methods; Fixman’s theorem; O(n); manipulator dynamics; revolute manipulators;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Robotics and Automation, 2005. ICRA 2005. Proceedings of the 2005 IEEE International Conference on
  • Print_ISBN
    0-7803-8914-X
  • Type

    conf

  • DOI
    10.1109/ROBOT.2005.1570849
  • Filename
    1570849