Enhanced vibration control of a multilink flexible manipulator using filtered inverse controller

Abstract

Flexible manipulators have numerous advantages such as lightweight, high operation speed, and low power consumption. However, they suffer from link vibrations, especially when operated at high speeds followed by sudden stops. This limitation has been addressed using techniques such as adaptive filters, adaptive strain feedback gain, state feedback control, etc. This article presents a filtered inverse controller for the mitigation of link vibrations in a multi-link flexible manipulator. To this end, the plant model, developed and linearized in Maple/Maplesim was inverted in MATLAB. The internal dynamics of the inverse model were stabilized using the state feedback technique. For safe and high-speed operations, the inverse model was augmented with a low pass filter to form the filtered inverse which was used as feedforward controller. Practical experiments were carried out in the dSPACE environment. Results show that filtered inverse controller yield not only faster response but relatively minimal link vibration when compared with the manipulator without vibration controller.