A Variable Structure-Based Estimation Strategy Applied to an RRR Robot System

Abstract

Nonlinear estimation strategies are important for accurate and reliable control of robotic manipulators. This brief paper studies the application of estimation theory to a simple robotic manipulator. Two estimation techniques are considered: the classic extended Kalman filter (EKF), and the robust smooth variable structure filter (SVSF). The EKF is included to present a basic background in estimation techniques and the SVSF is described and implemented on the system. We simulate the SVSF applied to a dynamically modeled three-link robotic manipulator. The results of the paper demonstrate that nonlinear estimation techniques such as the SVSF can be applied effectively to robots with modeling uncertainty and external disturbances.