Regarding human factors research on man-machine interface, a basic psychological experiment was conducted to observe psycho-physiological characteristics of on-line human cognitive behavior when cognitive tasks on learning and pattern classification were given to subjects by personal computer using a simple state transition model. During the experiment, three different types of subjects' data were recorded : (i) eye movement data by eye mark recorder, (ii) physio-electric signals by polygraph and (iii) verbal reports. Those subjects' data were analyzed with respect to : (i) the related human cognitive characteristics concerning problem solving strategy, measures of problem difficulty and mental image effect, (ii) observed eye movement characteristics such as saccade, attention, pupil reaction and blinking, etc., and (iii) obtained characteristics of skin potential response and heart rate. It was found that the application of psycho-physiological measurement would serve to objective and detailed analysis of on-line cognitive process.
In our previous paper 4) -12), we have studied “Theory of Invariance” for nonlinear continuous-time systems with their inputs appearing linearly. The aim of this paper is to derive necessary and sufficient conditions that inputs never effect outputs (so called conditions of “Invariance”) for nonlinear discrete-time and continuous-time systems without the assumption of linearity in inputs. Applying these conditions to the decoupling problem, we obtained necessary and sufficient conditions which yield decoupling control laws for both the continuous-time systems and the discrete-time systems. The conditions of decoupling given by van der Schaft 13) and Grizzle 14) via geometric approach are different in forms from ours. But, it seems difficult to derive control laws from their conditions because the integration of partial deferential equations is needed there.
This paper implements a knowledge-acquisition system for building a diagnostic knowledge-base of large-scale processing plants. The large-scale plants are so complicated to analyze numerically, further in some cases it is impossible to formulate the numerical models expressed in term of the differential equations. In view of this circumstance, a great use is made of the qualitative simulation to generate diagnostic rules. In researches about the qualitative simulation, the method of model description is concerned with knowledge explanation and the method for simulation. In this paper, we propose a qualitative network model in which an object-oriented approach is taken. The inheritance mechanism and the newly introduced instance linking mechanism allow systematic building of large-scale models.
Various continuous-time robust control methods have been proposed for nonlinear systems. However the implementation of these methods usually requires digital control systems. A digital implementation of a control algorithm based on a continuous-time theory often leads to chattering, even when a continuous input function is used, and makes the control error larger than the theoretically guaranteed control error limits. In this paper we propose a robust control scheme of nonlinear mechanical systems, on the premise of using the digital control. This scheme gives the admissible control error taking the sampling period into consideration, and introduces the weighting function for the input gain in order to reduce the chattering. Then it is shown that the proposed control scheme is effective by numerical simulations.