Distributed computation of inverse dynamics of robots

This paper presents a task scheduling to perform parallel computation of dynamics of robots. For illustration, the inverse dynamic analysis considered is based on the Newton-Euler recursive formulation. The results are presented for a robot with six degrees of freedom. The task scheduling has been prepared using the time taken to execute mathematical operations, transfer data between processors and perform substitutions for array operations. The paper reports a comparative study between T800 transputers and TMS320C40 parallel processors. It can be seen that the use of a single C40 processor results in 81% and faster performance compared to a single T800 while the use of two and three transputers provide a speed up of 28.8% and 43% respectively. Implementation of the task scheduling has been performed employing INMOS T800-20 transputers to evaluate the dynamics of Puma 560 robot without simplifying the general form of the equations employed