PID Control with Neural Compensation
Abstract
The simplest method to decrease the tracking error of PD control is to add an integral action, i.e., change the neural PD control into neural PID control. A natural question is: Why do we not add an integrator instead of increasing derivative gain in the neural PD control?
In this chapter, the well-known neural PD control of robot manipulators is extended to the neural PID control. It is an industrial linear PID controller adding a neural compensator. The semiglobal asymptotic stability of this novel neural control is proven. Explicit conditions for choosing PID gains are given. When the measurement of velocities it is not available, local asymptotic stability is also proven with a ...
Get PID Control with Intelligent Compensation for Exoskeleton Robots now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
