High-speed shutter function is indispensable to obtain a still image for the image processing of rapid moving objects. We propose a new video camera with the electronic shutter function using a frame transfer type CCD (FT-CD), and a new external trigger shutter function which enables to operate the shutter without transient image distortion at any external timing. The electronic shutter function is realized by such a method that stores necessary charges only for the duration of shutter timing by controlling directly FT-CD. As to “smear” which is likely to make a problem to FT-CCD, we confirmed by detailed analysis that “smear” provides no problem in practical uses. To prove the usefulness of this camera, we applied this to the vehicle-license number recognition unit and the automatic defect picture recording system of the paper manufacturing line. These results show that the use of this camera can be expanded to various fields.
All constituents in one real group sometimes do not have equal influence on group decision-making. Such is the case when a constituent gives another constituent full power of attorney on a blank sheet of paper. It is reasonable that collective choice rules with subordinate relations as the variables are used to such group decision-making. Then there sometimes exist subordinate relations, with which the influence of a constituent is connected such that its preference relation can be only reflected to the social preference relation in a collective choice rule. The relations are called the “winning subordinations” of the constituent in the rule. In this paper, properties of winning subordinations in two collective choice rules under majority rules are investigated when the subordinate relations are transitive. The subordinate degree of the influence is independent of the subordinate relations in one of the two rules. It is dependent of them in the other rule.
We propose a learning control scheme using a multi-layered neural network such that the control performance is improved for unknown controlled objects by repeated trials. This system was applied to the stabilization of an inverted pendulum. The system consists of three subsystems ; a neurocontroller, a temporary target generator and an error evaluator. The temporary target generator yields reference angle of the inverted pendulum to control the cart position. The error evaluator generates a teaching signal for the neuro controller by comparing the system error with an error-reference model. The backpropagation algorithm was used for the learning of neural network. The learning time could be shortened through the pre-learning process by imbedding a priori knowledge in the neural network. We verified the effectiveness of the proposed method by computer simulation.
Modal instability due to spillover is investigated for a flexible vibration control system that employs a modal filter as a state estimator. First the characteristic equation of the system equipped with actuators, sensors and a modal filter is derived in a determinantal form. Then it is shown that the characteristic equation can be expanded. By applying the perturbation technique to the expanded characteristic equation, a criterion is obtained to predict whether a residual mode become stable or unstable. Using the criterion, it is mathematically proved that the control system equipped with one actuator and one sensor always has unstable modes if they are non-colocated.
Disturbance instability in an adaptive D. C. servo system is investigated. The system considered in this paper is a simple positioning system with a D. C. servo motor. Although the conventional adaptive algorithm is adopted, complete stability analysis of the system under the existence of disturbances is difficult because of the nonlinearlity in the adaptive system. In this paper the drift of the adjustable parameters is discussed by singular perturbation method. Based on the analysis, some ideas to prevent the adjustable parameters from drifting are presented.
Parameters of the linear system in operation usually deviate from its nominal values used for the design. These deviations are generally time-varying and yield nonlinear effects even if the system is linear. Thus, it is difficult to deal with the time-varying parameter variations in the framework of linear system theory. In this paper, we give a variational expression for the output variation Δy (t) caused by the variable parameter term δμi (t) Dix (t) in the linear system. Using the expression, we can derive a necessary and/or sufficient condition for its first variation δy (t) to vanish in the open loop system, which are extended to the closed loop system with state feedback and reference input. This variational expression and derived conditions are useful for sensitivity analysis of structured parameter variations in linear systems.