Damping Oscillations in Material Handling Robot Using Input Shaping

When suspended payloads are moved in a material handling environment pendulum like oscillations are naturally introduced. This presents a problem any time especially when expensive and/or delicate objects are moved. In a steel industry environment objects have to be placed in tight locations after finishing up of rolling. Typically an object (rolled coil) is lifted by a hook on the end of a cable, creating pendulum that is free to swing in a transit condition. This swinging motion makes remote positioning of casks and barrels difficult to control precisely which is potentially destructive to facility equipment and to other storage containers. Typically, a crane operator moves objects slowly to minimize induced swinging and allow time for oscillation to dampen, maintaining safety but greatly decreasing the efficiency of operations. By employing proper algorithm or by the choice of proper drive it is possible to dampen the oscillations predictive called residual vibration[1]. Using damped-oscillation control algorithm [2] is one method of solving this problem. Residual vibrations resulting from the motion of the crane can not only be expensive to deal with but also dangerous. Standard practice in crane operation is to allow the pendulum swing to die out before working with the moved object. Considering the size of these cranes and the length of the pendulum, this could take a very long time. This situation lends itself to a more intelligent control system. Reducing these residual vibrations, even if not eliminating them, could save substantial wasted man-hours and create a safer working environment. This paper demonstrates the possibility of implementing an feedforward technique on a gantry crane robot model. Input shaping [3] is the technique of feedforward mode wherein the impulses are convolved with an arbitrary reference input to control the desired outputs or states of a system. This technique has been used in very limited situations. A Crane system is idealized as material handling industrial robot for experimentation. The model is tested under a steel environment using LABVIEW and PID algorithm