This paper deals with a nonlinear steady state modeling of river quality by a neural network which has a self-selection ability of network structure. This neural network algorithm is a revised one of the neural network which can identify a nonlinear system whose structure is very large and complex. By using measured data of river quality such as BOD and DO concentrations, a nonlinear steady state model of river quality is identified by the neural network and the results are compared with the results which are obtained by a physical model and a GMDH model. And it is shown that the neural network in this paper gives better prediction results as compared with a physical model and a GMDH model.
This paper describes a new binocular tracking method using Log Polar Mapping (LPM) which approximately represents the mapping of the retina into the visual cortex in primate vision : Using LPM makes it possible not only to obtain both a high central resolution and a wide field of view, but also to significantly reduce processing image data. In this paper, LPM is performed in software by lookup table method. Our tracking method utilizes zero disparity filter (ZDF) for extracting the target object and virtual horopter method for estimating binocular disparities, respectively. The performance of both target extraction and disparity estimation is improved in comparison with the conventional methods, by using LPM. Some experimental results are also shown to demonstrate the effectiveness of the proposed method.
This paper presents a two-degree freedom configuration of Generalized Predictive Control in order to increase the ability of the disturbance rejection and the setpoint tracking for a dead time system. The controller which is designed with the plant model including the feedforward control can prevent the undesirable action which is caused by the feedforward control. Numerical and experimental results demonstrate the effectiveness and the practicableness.
This paper presents a method of time series pattern recognition using recurrent neural networks (RNN) based on the genetic algorithms (GAs). The RNN could have dynamical characteristics because the RNN has feedback connections with time delay. The connection weights of the RNN transform into the gene which described by 16 bits binary code. In order to acquire the dynamics of the time series patterns, GAs operators which are a selection, a crossover and a mutation are applied to the gene of the RNN. In the numerical simulations, the training convergence of the RNN for some time series patterns are investigated. The simulation results show that the RNN evolved by the GAs has good performances of the training for time series patterns which generated from sine functions and for some pulse patterns. Finally, it is shown that the new method could determine some suitable network structures for time series training.
This paper is concerned with an application of the use of freedom in the coordinates transformation for exact linearizaion. We apply this technique to obstacle avoidance of a manipulator in the master-slave control. We show how to use the freedom for such an object and give the way to construct the linearizing coordinates transformation and feedback. And we evaluate the effectiveness of our technique by experiments.
This paper presents a genetic algorithm (GA) for scheduling problem of a robot control computation. That is very difficult problem which belongs to the class of NP-hard problems. The authors have already proposed several algorithms, which are based on heuristic and branch-and-bound approaches, for the scheduling problem. The conventional algorithms, however, have the limits of their ability in quality of solutions and computational time. The scheduling problem is a typical partitioning problem : partitioning objects into a fixed number of groups to optimize an objective function. Consequently, this paper proposes a new crossover method named weighted-edge crossover which preserves both the structure and the characteristic of the feasible solution of partitioning problem. Furthermore, in order to improve the performance of GA, this paper defines a distance between feasible solutions and uses it in the adaptive control of crossover rate. To demonstrate the effectiveness of the proposed GA, comparative study of the GA with the conventional algorithms is carried out on several computational experiments.
The uncertainty model used in the H∞ control is norm bounded, while the actual modeling error is not so in general, and hence such uncertainty model can be conservative in certain applications. Therefore, the resulting system performance might be conservative if we simply apply the H∞ control scheme. An H∞ control scheme with minor feedback is examined to reduce the conservativeness. Since the minor feedback loop changes both the nominal plant and the uncertainty models, we cannot conclude immediately that it improves system performances even if the magnitude of uncertainty model is reduced. A disturbance attenuation problem is discussed to investigate the effect of the minor feedback, and a sufficient condition is derived on which the minor feedback improves system performance. Experimental results for a pneumatic pressure control system confirm the effectiveness of the proposed method.