A CORDIC-Based Pipelined Architecture
暂无分享,去创建一个
This paper describes a novel CORDIC-based pipelined architecture for computing direct kinematic position solution based on the decomposition of the homogeneous link transformation matrix. It is found that a homogeneous link transformation matrix can be decomposed into a product of two matrices, each of which can be computed by two CORDIC processors arranged in parallel, forming a 2-stage cascade CORDIC computational module. Extending this idea to an n-link manipulator, n 2-stage CORDIC computational modules, consisting of 4n CORDIC processors, can be concatenated to form a pipelined architecture for computing the position and orientation of the end-effector of the manipulator. Since the initial delay time of the proposed pipelined architecture is 80np and the pipelined time is 40p. the proposed CORDICbased architecture requires a total computation time of (80n + 120)p for computing the position and orientation of the end-effector of an n -link manipulator.
[1] J. Denavit,et al. A kinematic notation for lower pair mechanisms based on matrices , 1955 .
[2] Jean-Marc Delosme,et al. Highly concurrent computing structures for matrix arithmetic and signal processing , 1982, Computer.