Implementation of a MRAC for a Two Axis Direct Drive Robot Manipulator using a Digital Signal Processor

This paper is concerned with the digital implementation of a Model Reference Adaptive Control (MRAC) algorithm on a Texas Instrument TMS32010 Digital Signal Processor (DSP). The MRAC was designed to control a two axis direct drive SCARA type robot manipulator. The primary purpose of the adaptive controller is to compensate for the inertial variations due to changes in configurtion and payload. Experimental results presented clearly illustrate the need for adaptive control over conventional PID controller for the type of arm structure used in the experiments. Discussion on the use of DSP in controls is presented in terms of their capabilities and the influence their performance will have on the sampling time of digital control systems.