Compression for computation delays in leader-follower coordination of two manipulators using exponential time functions in joint space

Coordinated control of two manipulators using discrete position feedback is developed and experimentally tested using a hierarchical multiprocessor architecture and two PUMA robot arms. Since the research is heavily experimental, leader-follower coordination is adopted because of its computational efficiency. The computation of desired follower arm positions consumes one sampling interval. Another sampling interval is lost due to the fact that a desired follower arm position has to be computed before it can be assumed. The design of a class of prediction equations to compensate for these timing delays is emphasized. These equations are developed by computing an estimated model for the delayed desired follower arm trajectory using exponential time functions. Both exact and least squares spline fitting are used and compared.<>