Restricted Structure Non-Linear Generalized Minimum Variance Control of a 2-Link Robot Arm

Abstract

The objective of this paper is to propose a restricted structure non-linear generalized minimum variance (RS-NGMV) controller for a two-link robot arm. The NGMV control is a useful method for offering control solutions for nonlinear systems. The motivation is to provide the advantages of NGMV control inside a low-order controller structure with an intention to enable design simplicity and easy implementation for engineers with classical training. The result will be an optimal controller with simple tuning variables. Simulations of the RS-NGMV controller are presented using Matlab/Simulink.

Speed Tracking of an Electric Vehicle Using a Restricted Structure NGMV Control Algorithm

“…1. The controller is a special case, referred to as the parallel form RS-NGMV [4,5], which is useful if a low-order controller that already stabilizes the system is available. The parallel form control input signal is a combination of fixed gain RS-PI given in (27) and the gain deviations component, .…”

“…The RS-NGMV algorithm can use a state-space based Non-linear Generalized Minimum Variance (NGMV) control solution but given a multi-variable Restricted-Structure (RS) controller that is in a low order z-transfer function form within the feedback loop. The aim is to make the controller within the loop have a familiar form for ease of tuning and a classical low order controller such as PID [4,5] may be used. The low-order restricted structure controller has background processing that has a performance similar to the advanced model based NGMV control solution.…”