A VLSI-based robot dynamics learning algorithm

A computationally efficient solution to the problem of identifying the dynamic parameters of a robot manipulator is presented. The method is based on a simplified representation of the dynamics based on the Lagrange-Euler formulation. For parameter estimation, a recursive least squares technique is used. The approach has three important characteristics: (1) since they are based on the Lagrangian representation, the equations are linear in the dynamic parameters, enabling the application of linear identification techniques; (2) the dynamic parameters are easily recognized, extracted, and grouped; and (3) the equations are amenable to the implementation of parallel processing schemes. The algorithm was distributed over a network of transputers. Real-time results have been produced to demonstrate the speedup and efficiency of the proposed technique