Optimal input shaping for vibration control of a flexible manipulator using GA

Abstract

This paper presents optimization of input shaping technique using genetic algorithms (GAs) for a planar single link flexible manipulator system with varying payload. An unshaped bang-bang torque is shaped by convolving a sequence of impulses with an input shaper. The time locations of the impulses are determined using GA. The simulation work is designed in Matlab based environment and the responses are presented in the time and frequency domains. The performance of optimal input shaper is compared to the bang-bang torque in terms of maneuver speed, transient and steady state responses, computational complexity and level of reduction at resonances mode.