Masaki Odai

Dual rate Kalman filter considering delayed measurement and its application in visual servo

This paper describes a Kalman filter for a class of linear system in which: 1) The sampling time of measurement is longer than the control period. 2) The measurement is delayed. Using the delayed measurement and the present and past time predicted state, the pseudo-measurement in present time is constructed. The optimal Kalman gain is designed to minimize the trace of the covariance of estimation error. A dual rate recursive algorithm is proposed to estimate the state synchronously with the control signal. A visual servo system is demonstrated to verify the effectiveness of the proposed method.

Fast and accurate vision-based positioning control employing multi-rate Kalman filter

To achieve fast and accurate positioning control, a controller based on a visual servo employing a multi-rate Kalman filter was implemented and experimentally evaluated. Although a visual servo is expected to achieve robust positioning control by eliminating positional errors due to operation environments, applying visual information to positioning control faces three challenges, namely, lower sampling speed, longer delay, and lower positional accuracy than those possible by using a conventional position encoder. To overcome these problems, a multi-rate Kalman filter with a time-delay compensation technique is employed to combine visual information with encoder-based positional information. Two models describing a positioning-control experimental system using a one-axis linear stage with a camera were investigated to determine the optimum structure of the Kalman filter. Evaluation of the experimental system showed that the designed controller achieves high-precision (±10 μm) positioning control within approximately 10-ms settling time while simultaneously adjusting the environmental positional errors.

Evaluation of positioning control system using relative displacement observer for precise positioning stages

Precise positioning technologies are widely used for industrial machines such as device manufacturing equipment. There have been many studies on state feedback control systems for precise positioning stages that improve accuracy by using sensor feedback. However, the sensor information cannot always indicate the performance of industrial machines directly because the final requirement of a stage system is generally to adjust a movable table to a target structure. Therefore, the relative relationship between the stage and the target structure must be controlled. However, the stage and other structures must be considered to be elastic elements, especially for precise positioning devices. Thus, this study aims to achieve high-precision relative positioning between flexible structures. The authors had proposed a relative displacement observer which estimated the relative displacement between the stage and the target structure by using only the stage displacement. However, in industrial applications, robustness against disturbances such as modeling errors and aging changes are required. In this paper, we show evaluation results of robustness against accuracy of model identification and aging change in resonant frequency of plant. Results of a numerical simulation show that the proposed control system improves relative positioning performance. And also, we confirm that it has enough robustness against aging change of the stage system.