In this investigation, a new algorithm is developed for the updating of a neural network. It is concentrated in a fuzzy transition between the recursive least square and extended Kalman filter algorithms with the purpose to get a bounded gain such that a satisfactory modeling could be maintained. The advised algorithm has the advantage compared with the mentioned methods that it eludes the excessive increasing or decreasing of its gain. The gain of the recommended algorithm is uniformly stable and its convergence is found. The new algorithm is employed for the modeling of two synthetic examples.
This paper considers the data-based identification of industrial robots using an instrumental variable method that uses off-line estimation of the joint velocities and acceleration signals based only on the measurement of the joint positions. The usual approach to this problem relies on a `tailor-made' prefiltering procedure for estimating the derivatives that depends on good prior knowledge of the system's bandwidth. The paper describes an alternative Integrated Random Walk SMoothing (IRWSM) method that is more robust to deficiencies in such a priori knowledge and exploits an optimal recursive algorithm based on a simple integrated random walk model and a Kalman filter with associated fixed interval smoothing. The resultant IDIM-IV instrumental variable method, using this approach to signal generation, is evaluated by its application to an industrial robot arm and comparison with previously proposed methods.
Considering the correlations of the input indexes and the deficiency of calibrating kernel function parameters when support vector machine (SVM) is applied, a forecasting method based on principal component analysis-genetic algorithm-support vector machine (PCA-GA-SVM) is proposed to improve the precision of bus arrival time prediction. And the No. 232 bus in Shenyang City of China is taken as an example. The traditional SVM and Kalman Filtering model and GA-SVM are also employed to make comparative analysis on the prediction rate, respectively. The result indicates that PCA-GA-SVM obtains more accurate prediction results of bus arrival time prediction.
Resolution of seamless switching within a set of available wireless access solutions applied in ITS (Intelligent Transport System) applications is presented. The CALM based system or specifically designed and configured L3/L2 switching can be an adequate solution for the multi-path access communication system. These systems have to meet requirements of the seamless secure communications functionality within even an extensive cluster of moving objects. Competent decision processes based on precisely quantified system requirements and each performance indicator tolerance range must be implemented to keep service up and running with no influence of continuously changing conditions in time and served space.
The method of different paths service quality evaluation and selection of the best possible active communications access path is introduced. This method is studied within project STATVU (System Requirements and Architecture of the Universal Telematic Vehicle Unit is grant 2A-1TP1/138 of the Ministry of Industry and Trade of the Czech Republic.). The proposed approach is based on Kalman filtering which separates reasonable part of noise and also allows prediction of the individual parameters near future behavior. The presented classification algorithm applied on filtered measured data combined with deterministic parameters is trained using historical data, i. e. combination of parameters vectors line and relevant decisions. Quality of classification is dependent on the size and quality of the training data sets.
Recursive time series methods are very popular due to their numerical simplicity. Their theoretical background is usually based on Kalman filtering in state space models (mostly in dynamic linear systems). However, in time series practice one must face frequently to outlying values (outliers), which require applying special methods of robust statistics. In the paper a simple robustification of Kalman filter is suggested using a simple truncation of the recursive residuals. Then this concept is applied mainly to various types of exponential smoothing (recursive estimation in Box-Jenkins models with outliers is also mentioned). The methods are demonstrated using simulated data.
The robust recursive algorithm for the parameter estimation and the volatility prediction in GARCH models is suggested. It seems to be useful for various financial time series, in particular for (high-frequency) log returns contaminated by additive outliers. The proposed procedure can be effective in the risk control and regulation when the prediction of volatility is the main concern since it is capable to distinguish and correct outlaid bursts of volatility. This conclusion is demonstrated by simulations and real data examples presented in the paper.