Aim To propose a modelling method for flexible manipulators. Methods The improved algorithm and structure of the ANN (artificial neural networks) were used. All of the data used in the process of modelling came from e...Aim To propose a modelling method for flexible manipulators. Methods The improved algorithm and structure of the ANN (artificial neural networks) were used. All of the data used in the process of modelling came from experiments based on a very flexible link which was fixed on a FANUC Robot S-Model 300 in our lab.Results and Conclusion The theoretical analysis and experiment results showed that this modelling scheme is more suitable for flexible systems with characteristics of fast changing dynamics, and also it can be more accurate than others and is more convenient for real-time use.展开更多
The non-linear relationships between the contents of ginsenoside Rg 1, Rb 1, Rd and Panax notoginseng saponins(PNS) in Panax notoginseng root herb and the near infrared(NIR) diffuse reflectance spectra of the herb wer...The non-linear relationships between the contents of ginsenoside Rg 1, Rb 1, Rd and Panax notoginseng saponins(PNS) in Panax notoginseng root herb and the near infrared(NIR) diffuse reflectance spectra of the herb were established by means of artificial neural networks(ANNs). Four three-layered perception feed-forward networks were trained with an error back-propagation algorithm. The significant principal components of the NIR spectral data matrix were utilized as the input of the networks. The networks architecture and parameters were selected so as to offer less prediction errors. Relative prediction errors for Rg 1, Rb 1, Rd and PNS obtained with the optimum ANN models were 8.99%, 6.54%, 8.29%, and 5.17%, respectively, which were superior to those obtained with PLSR methods. It is verified that ANN is a suitable approach to model this complex non-linearity. The developed method is fast, non-destructive and accurate and it provides a new efficient approach for determining the active components in the complex system of natural herbs.展开更多
Consider the design and implementation of an electro-hydraulic control system for a robotic excavator, namely the Lancaster University computerized and intelligent excavator (LUCIE). The excavator was developed to aut...Consider the design and implementation of an electro-hydraulic control system for a robotic excavator, namely the Lancaster University computerized and intelligent excavator (LUCIE). The excavator was developed to autonomously dig trenches without human intervention. One stumbling block is the achievement of adequate, accurate, quick and smooth movement under automatic control, which is difficult for traditional control algorithm, e.g. PI/PID. A gain scheduling design, based on the true digital proportional-integral-plus (PIP) control methodology, was utilized to regulate the nonlinear joint dynamics. Simulation and initial field tests both demonstrated the feasibility and robustness of proposed technique to the uncertainties of parameters, time delay and load disturbances, with the excavator arm directed along specified trajectories in a smooth, fast and accurate manner. The tracking error magnitudes for oblique straight line and horizontal straight line are less than 20 mm and 50 mm, respectively, while the velocity reaches 9 m/min.展开更多
A new class of support vector machine, nil-support vector machine, isdiscussed which can handle both classification and regression. We focus on nu-support vector machineregression and use it for phase space prediction...A new class of support vector machine, nil-support vector machine, isdiscussed which can handle both classification and regression. We focus on nu-support vector machineregression and use it for phase space prediction of chaotic time series. The effectiveness of themethod is demonstrated by applying it to the Henon map. This study also compares nu-support vectormachine with back propagation (BP) networks in order to better evaluate the performance of theproposed methods. The experimental results show that the nu-support vector machine regressionobtains lower root mean squared error than the BP networks and provides an accurate chaotic timeseries prediction. These results can be attributable to the fact that nu-support vector machineimplements the structural risk minimization principle and this leads to better generalization thanthe BP networks.展开更多
Inspired by the traditional Wold's nonlinear PLS algorithm comprises of NIPALS approach and a spline inner function model,a novel nonlinear partial least squares algorithm based on spline kernel(named SK-PLS)is pr...Inspired by the traditional Wold's nonlinear PLS algorithm comprises of NIPALS approach and a spline inner function model,a novel nonlinear partial least squares algorithm based on spline kernel(named SK-PLS)is proposed for nonlinear modeling in the presence of multicollinearity.Based on the inner-product kernel spanned by the spline basis functions with infinite number of nodes,this method firstly maps the input data into a high-dimensional feature space,and then calculates a linear PLS model with reformed NIPALS procedure in the feature space and gives a unified framework of traditional PLS "kernel" algorithms in consequence.The linear PLS in the feature space corresponds to a nonlinear PLS in the original input(primal)space.The good approximating property of spline kernel function enhances the generalization ability of the novel model,and two numerical experiments are given to illustrate the feasibility of the proposed method.展开更多
A new algorithm of nonuniformity correction for infrared focal plane array(IRFPA) is reported,which is a combined algorithm based on both the two-point correction and artificial neural networks correction. The combine...A new algorithm of nonuniformity correction for infrared focal plane array(IRFPA) is reported,which is a combined algorithm based on both the two-point correction and artificial neural networks correction. The combined algorithm is calibrated by two-point correction,and the calibrated correction coefficients are automatically modified by BP algorithm. So it is not only calibrated,but also real-time processed. In adaptive nonuniformity correction algorithm,the phenomena ghost artifact and target fade-out are avoided by edge extraction. In order to get intensified image,the modified median filters are adopted. The simulated data indicates the proposed scheme is an effective algorithm.展开更多
文摘Aim To propose a modelling method for flexible manipulators. Methods The improved algorithm and structure of the ANN (artificial neural networks) were used. All of the data used in the process of modelling came from experiments based on a very flexible link which was fixed on a FANUC Robot S-Model 300 in our lab.Results and Conclusion The theoretical analysis and experiment results showed that this modelling scheme is more suitable for flexible systems with characteristics of fast changing dynamics, and also it can be more accurate than others and is more convenient for real-time use.
文摘The non-linear relationships between the contents of ginsenoside Rg 1, Rb 1, Rd and Panax notoginseng saponins(PNS) in Panax notoginseng root herb and the near infrared(NIR) diffuse reflectance spectra of the herb were established by means of artificial neural networks(ANNs). Four three-layered perception feed-forward networks were trained with an error back-propagation algorithm. The significant principal components of the NIR spectral data matrix were utilized as the input of the networks. The networks architecture and parameters were selected so as to offer less prediction errors. Relative prediction errors for Rg 1, Rb 1, Rd and PNS obtained with the optimum ANN models were 8.99%, 6.54%, 8.29%, and 5.17%, respectively, which were superior to those obtained with PLSR methods. It is verified that ANN is a suitable approach to model this complex non-linearity. The developed method is fast, non-destructive and accurate and it provides a new efficient approach for determining the active components in the complex system of natural herbs.
基金Project(K5117827)supported by Scientific Research Foundation for the Returned Overseas Chinese ScholarsProject(08KJB510021)supported by the Natural Science Research Council of Jiangsu Province,China+1 种基金Project(Q3117918)supported by Scientific Research Foundation for Young Teachers of Soochow University,ChinaProject(60910001)supported by National Natural Science Foundation of China
文摘Consider the design and implementation of an electro-hydraulic control system for a robotic excavator, namely the Lancaster University computerized and intelligent excavator (LUCIE). The excavator was developed to autonomously dig trenches without human intervention. One stumbling block is the achievement of adequate, accurate, quick and smooth movement under automatic control, which is difficult for traditional control algorithm, e.g. PI/PID. A gain scheduling design, based on the true digital proportional-integral-plus (PIP) control methodology, was utilized to regulate the nonlinear joint dynamics. Simulation and initial field tests both demonstrated the feasibility and robustness of proposed technique to the uncertainties of parameters, time delay and load disturbances, with the excavator arm directed along specified trajectories in a smooth, fast and accurate manner. The tracking error magnitudes for oblique straight line and horizontal straight line are less than 20 mm and 50 mm, respectively, while the velocity reaches 9 m/min.
文摘A new class of support vector machine, nil-support vector machine, isdiscussed which can handle both classification and regression. We focus on nu-support vector machineregression and use it for phase space prediction of chaotic time series. The effectiveness of themethod is demonstrated by applying it to the Henon map. This study also compares nu-support vectormachine with back propagation (BP) networks in order to better evaluate the performance of theproposed methods. The experimental results show that the nu-support vector machine regressionobtains lower root mean squared error than the BP networks and provides an accurate chaotic timeseries prediction. These results can be attributable to the fact that nu-support vector machineimplements the structural risk minimization principle and this leads to better generalization thanthe BP networks.
文摘Inspired by the traditional Wold's nonlinear PLS algorithm comprises of NIPALS approach and a spline inner function model,a novel nonlinear partial least squares algorithm based on spline kernel(named SK-PLS)is proposed for nonlinear modeling in the presence of multicollinearity.Based on the inner-product kernel spanned by the spline basis functions with infinite number of nodes,this method firstly maps the input data into a high-dimensional feature space,and then calculates a linear PLS model with reformed NIPALS procedure in the feature space and gives a unified framework of traditional PLS "kernel" algorithms in consequence.The linear PLS in the feature space corresponds to a nonlinear PLS in the original input(primal)space.The good approximating property of spline kernel function enhances the generalization ability of the novel model,and two numerical experiments are given to illustrate the feasibility of the proposed method.
文摘A new algorithm of nonuniformity correction for infrared focal plane array(IRFPA) is reported,which is a combined algorithm based on both the two-point correction and artificial neural networks correction. The combined algorithm is calibrated by two-point correction,and the calibrated correction coefficients are automatically modified by BP algorithm. So it is not only calibrated,but also real-time processed. In adaptive nonuniformity correction algorithm,the phenomena ghost artifact and target fade-out are avoided by edge extraction. In order to get intensified image,the modified median filters are adopted. The simulated data indicates the proposed scheme is an effective algorithm.
