OPTIMAL PERFORMANCE IN A STEP MOTOR CONTROLLED ROBOT.

摘要

A step motor equipped robot is studied mainly from the control point of view. Step motors are used widely in industry, mostly for position control purposes. As an actuator it combines qualities which make it attractive for robotics uses. It is a digital device, does not need extra measurements for position and velocity, such as shaft encoders, and it can be easily braked. In general, on the system level a step motor equipped robot requires less computer power and fewer hardware devices.; In order to make use of these advantages, a mechanical manipulator without feedback was chosen to be the model for this research, and as a result, preview control calculations were necessary.; The 'Maximum principle' theorem was used to design an optimal control based on minimum time for a specified motion. With the use of an optimization technique the optimal control problem was solved numerically using the complete, highly nonlinear dynamic model of the manipulator.; In this solution, system boundaries and obstacles inside the working space are taken into account.; The output results are velocities and position states which guarantee that the maximum available torques will be used in such a way that the manipulator will be maneuvered to the new position in a minimum time.; To implement these ideas, a six degree of freedom step motor driven robot was built and tested successfully in a real time control operation.; The conclusions of this work are: step motors can be used successfully for motion control as well as for position purposes. A robot equipped with step motors can provide a good performance while keeping the control system very simple.; The optimal control scheme can be applied to any type of manipulator with any actuator. It can handle the Coriolis and centripetal terms, Coulomb friction, and gear backlash, which are normally not easy to treat in nonlinear feedback control systems. With the new generation of microcomputers coming up with a fast floating point processor, the optimal control scheme can be solved off line but on the same computer that implements the real time control package. Based on this work time varying obstacles cases in robotics can be worked out; cases such as two or more manipulators cooperating in an optimal way in the same working space.