| Current approaches for high speed color printers require sheets be accurately positioned as they arrive to the image transfer station (ITS). This goal has been achieved by designing and building a steerable nips mechanism, which is located upstream from the ITS. This mechanism consists of two rollers, which not only rotate to advance the paper along the track, but also steer the paper in the yaw direction. Two servo motors are used for the rotation of the rollers and two other servo motors are used to steer them. In this dissertation we will present the design, experimental setup, system model, and the observer and control strategies needed to precisely correct for the lateral and angular positions of the sheet as well as to deliver it on time to the ITS. In addition, since the two rollers can steer independently from each other, the amount of sheet buckling and/or stretching must also be controlled and constrained within acceptable limits to prevent the sheet from getting damaged.;The steerable nips system model is nonlinear and subject to four nonholonomic constraints. In order to control the position and orientation of the sheet, we developed a robust approach to dynamic feedback linearization by feedback linearizing only the kinematics of the system and using two internal loops for the local control of the actuators' velocities and positions. The stability and robustness to unmodeled actuator dynamics of the control system are also proven in this dissertation.;This control strategy was successfully implemented in the experimental setup using a sensing scheme based on five photodiode sensors used for the detection of the leading edge of the sheet, and two laser sensors used for the detection of the lateral edge of the sheet. Whereas the angular orientation of the page is measured directly by the two laser sensors, the other sheet states (lateral and longitudinal position as well as sheet buckling) need to be estimated when the leading edge of the sheet is in between photodiodes. An open loop observer, based on the system kinematics, is used for such estimation.;This thesis also developed a closed-loop nonlinear observer and describes its implementation on the experimental setup. Such observer is based on the Extended Luenberger Observer for MIMO systems, which is an extension to the well known Normal Form observer. In order to estimate all sheet states, this nonlinear observer requires only one laser sensor for lateral sheet detection and continuous measurements for the leading edge of the sheet. |
