Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617394
Liviu-Cristian Duţu, G. Mauris, P. Bolon, Stéphanie Dabic, J. Tissot
Nowadays tactile surfaces are slowly replacing the mechanical interfaces of our electronic devices, and the actual trend is toward a quasi-total touch interaction. This transition has however one important side effect, i.e the lack of feedback from the device, which in certain situations can be crucial. In order to overcome this, it has been suggested that feedback has to be delivered to the finger through vibrations that should be both detectable and comfortable. This paper aims to define a perception model for the sensory evaluations of the vibrotactile signals using fuzzy set theory. First of all, the hypothesis that haptic perception is strongly related to physical characteristics of the signals was evaluated and confirmed with a 93% correlation rate, based on psychophysical studies of the tactile sense. Secondly, using the previous analysis as a knowledge base we have implemented a fuzzy inference system which forecasts the preference values for vibrotactile signals. The preliminary results show that for 15 out of 18 signals, the preference is correctly predicted within a reasonable uncertainty interval.
{"title":"A fuzzy model relating vibrotactile signal characteristics to haptic sensory evaluations","authors":"Liviu-Cristian Duţu, G. Mauris, P. Bolon, Stéphanie Dabic, J. Tissot","doi":"10.1109/CIVEMSA.2013.6617394","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617394","url":null,"abstract":"Nowadays tactile surfaces are slowly replacing the mechanical interfaces of our electronic devices, and the actual trend is toward a quasi-total touch interaction. This transition has however one important side effect, i.e the lack of feedback from the device, which in certain situations can be crucial. In order to overcome this, it has been suggested that feedback has to be delivered to the finger through vibrations that should be both detectable and comfortable. This paper aims to define a perception model for the sensory evaluations of the vibrotactile signals using fuzzy set theory. First of all, the hypothesis that haptic perception is strongly related to physical characteristics of the signals was evaluated and confirmed with a 93% correlation rate, based on psychophysical studies of the tactile sense. Secondly, using the previous analysis as a knowledge base we have implemented a fuzzy inference system which forecasts the preference values for vibrotactile signals. The preliminary results show that for 15 out of 18 signals, the preference is correctly predicted within a reasonable uncertainty interval.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125740729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617398
Yong Hu, Jerry Kwok, J. Tse
The contaminated electrocardiography (ECG) is a big problem in the surface electromyography (SEMG) signal detection and analysis. The objective of the current study is to propose and validate an algorithm for the automated feature cognition and identification for eliminating ECG artifact from the raw SEMG signals. The utilization of Independent Component Analysis (ICA) method is to decompose the raw SEMG signals into individual independent source components. After that, some of the independent source components with the characteristics of ECG artifact were detected by the automated identification algorithm and thereafter eliminated. The sensitivity and specificity of the algorithm for distinguishing ECG source components from independent source components are 100% and 99% respectively. The automated identification algorithm exhibits the prominent performance of recognition for ECG artifact and can be considered reliable and effective.
{"title":"Automatic ECG artifact removal in the real-time SEMG recording system","authors":"Yong Hu, Jerry Kwok, J. Tse","doi":"10.1109/CIVEMSA.2013.6617398","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617398","url":null,"abstract":"The contaminated electrocardiography (ECG) is a big problem in the surface electromyography (SEMG) signal detection and analysis. The objective of the current study is to propose and validate an algorithm for the automated feature cognition and identification for eliminating ECG artifact from the raw SEMG signals. The utilization of Independent Component Analysis (ICA) method is to decompose the raw SEMG signals into individual independent source components. After that, some of the independent source components with the characteristics of ECG artifact were detected by the automated identification algorithm and thereafter eliminated. The sensitivity and specificity of the algorithm for distinguishing ECG source components from independent source components are 100% and 99% respectively. The automated identification algorithm exhibits the prominent performance of recognition for ECG artifact and can be considered reliable and effective.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117046089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617388
D. Ionescu, V. Suse, C. Gadea, B. Solomon, B. Ionescu, S. Islam
Gesture Control dominates presently the research on new human computer interfaces. The domain covers both the sensors to capture gestures and also the driver software which interprets the gesture mapping it onto a robust command. More recently, there is a trend to use depth-mapping camera as the 2D cameras fall short in assuring the conditions of real-time robustness of the whole system. As image processing is at the core of the detection, recognition, and tracking the gesture, depth mapping sensors have to provide a depth image insensitive to illumination conditions. Thus depth-mapping cameras work in a certain wavelength of the infrared (IR) spectrum. In this paper, a novel real-time depth-mapping principle for an IR camera is introduced. The new IR camera architecture comprises an illuminator module which is pulse-modulated via a monotonic function using a cycle driven feedback loop for the control of laser intensity, while the reflected infrared light is captured in “slices” of the space in which the object of interest is situated. A reconfigurable hardware architecture unit calculates the depth slices and combines them in a depth-map of the object to be further used in the detection, tracking, and recognition of the gesture made by the user. Images of real objects are reconstructed in 3D based on the data obtained by the space-slicing technique, and a corresponding image processing algorithm builds the 3D map of the object in real-time. As this paper will show through a series of experiments, the camera can be used in a variety of domains, including for gesture control of 3D objects in virtual environments.
{"title":"An infrared-based depth camera for gesture-based control of virtual environments","authors":"D. Ionescu, V. Suse, C. Gadea, B. Solomon, B. Ionescu, S. Islam","doi":"10.1109/CIVEMSA.2013.6617388","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617388","url":null,"abstract":"Gesture Control dominates presently the research on new human computer interfaces. The domain covers both the sensors to capture gestures and also the driver software which interprets the gesture mapping it onto a robust command. More recently, there is a trend to use depth-mapping camera as the 2D cameras fall short in assuring the conditions of real-time robustness of the whole system. As image processing is at the core of the detection, recognition, and tracking the gesture, depth mapping sensors have to provide a depth image insensitive to illumination conditions. Thus depth-mapping cameras work in a certain wavelength of the infrared (IR) spectrum. In this paper, a novel real-time depth-mapping principle for an IR camera is introduced. The new IR camera architecture comprises an illuminator module which is pulse-modulated via a monotonic function using a cycle driven feedback loop for the control of laser intensity, while the reflected infrared light is captured in “slices” of the space in which the object of interest is situated. A reconfigurable hardware architecture unit calculates the depth slices and combines them in a depth-map of the object to be further used in the detection, tracking, and recognition of the gesture made by the user. Images of real objects are reconstructed in 3D based on the data obtained by the space-slicing technique, and a corresponding image processing algorithm builds the 3D map of the object in real-time. As this paper will show through a series of experiments, the camera can be used in a variety of domains, including for gesture control of 3D objects in virtual environments.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127194357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617400
C. Roadknight, U. Aickelin, J. Scholefield, L. Durrant
In this paper, we describe a dataset relating to cellular and physical conditions of patients who are operated upon to remove colorectal tumours. This data provides a unique insight into immunological status at the point of tumour removal, tumour classification and post-operative survival. We build on existing research on clustering and machine learning facets of this data to demonstrate a role for an ensemble approach to highlighting patients with clearer prognosis parameters. Results for survival prediction using 3 different approaches are shown for a subset of the data which is most difficult to model. The performance of each model individually is compared with subsets of the data where some agreement is reached for multiple models. Significant improvements in model accuracy on an unseen test set can be achieved for patients where agreement between models is achieved.
{"title":"Ensemble learning of colorectal cancer survival rates","authors":"C. Roadknight, U. Aickelin, J. Scholefield, L. Durrant","doi":"10.1109/CIVEMSA.2013.6617400","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617400","url":null,"abstract":"In this paper, we describe a dataset relating to cellular and physical conditions of patients who are operated upon to remove colorectal tumours. This data provides a unique insight into immunological status at the point of tumour removal, tumour classification and post-operative survival. We build on existing research on clustering and machine learning facets of this data to demonstrate a role for an ensemble approach to highlighting patients with clearer prognosis parameters. Results for survival prediction using 3 different approaches are shown for a subset of the data which is most difficult to model. The performance of each model individually is compared with subsets of the data where some agreement is reached for multiple models. Significant improvements in model accuracy on an unseen test set can be achieved for patients where agreement between models is achieved.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"168 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129987940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617414
R. D. Labati, A. Genovese, V. Piuri, F. Scotti
In this paper, we present a complete virtual environment for the computation of synthetic three-dimensional samples representing free falling wood strands.The proposed method permits to simulate acquisitions performed by real multiple view setups in which the stream of strands falling out of a conveyor belt is analyzed with image processing techniques in order to compute the particle size distribution. Unfortunately, experiments in real time applications are complex and expensive, and the ground true is almost impossible to measure in such conditions. The creation of a metric and fully virtual environment of falling wood strands represent a key feature in order to properly design the illuminotecnic and optical setups, optimize the image processing methods as well as the three- dimensional reconstruction techniques, using controlled and fully repeatable virtual image datasets.
{"title":"A virtual environment for the simulation of 3D wood strands in multiple view systems for the particle size measurements","authors":"R. D. Labati, A. Genovese, V. Piuri, F. Scotti","doi":"10.1109/CIVEMSA.2013.6617414","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617414","url":null,"abstract":"In this paper, we present a complete virtual environment for the computation of synthetic three-dimensional samples representing free falling wood strands.The proposed method permits to simulate acquisitions performed by real multiple view setups in which the stream of strands falling out of a conveyor belt is analyzed with image processing techniques in order to compute the particle size distribution. Unfortunately, experiments in real time applications are complex and expensive, and the ground true is almost impossible to measure in such conditions. The creation of a metric and fully virtual environment of falling wood strands represent a key feature in order to properly design the illuminotecnic and optical setups, optimize the image processing methods as well as the three- dimensional reconstruction techniques, using controlled and fully repeatable virtual image datasets.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123127915","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617417
M. Alreshoodi, J. Woods
A model that can predict an end user satisfaction or QoE (Quality of Experience) directly from the network QoS (Quality of Service) is still illusive in the field of image processing and is completely absent in multi-layered video. This motivates the derivation of a meaningful QoS to QoE mapping function to allow one to be predicted in the absence of the other. This paper presents an affine fuzzy logic based system that can map the QoS to QoE and can be extended to layered video streaming. The proposed methodology employs a learning system which optimizes the coded layered video for best QoE. Four QoS parameters are chosen as the inputs of the designed model, while the output is the Peak Signal-to-Noise Ratio (PSNR). The designed membership functions and the fuzzy rules extracted from the input and the output enable the proposed model to identify and learn the video QoE.
{"title":"An empirical study based on a fuzzy logic system to assess the QoS/QoE correlation for layered video streaming","authors":"M. Alreshoodi, J. Woods","doi":"10.1109/CIVEMSA.2013.6617417","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617417","url":null,"abstract":"A model that can predict an end user satisfaction or QoE (Quality of Experience) directly from the network QoS (Quality of Service) is still illusive in the field of image processing and is completely absent in multi-layered video. This motivates the derivation of a meaningful QoS to QoE mapping function to allow one to be predicted in the absence of the other. This paper presents an affine fuzzy logic based system that can map the QoS to QoE and can be extended to layered video streaming. The proposed methodology employs a learning system which optimizes the coded layered video for best QoE. Four QoS parameters are chosen as the inputs of the designed model, while the output is the Peak Signal-to-Noise Ratio (PSNR). The designed membership functions and the fuzzy rules extracted from the input and the output enable the proposed model to identify and learn the video QoE.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121191821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617395
Lucas Silva, R. Dantas, Andre L. H. Pantoja, Antonio Pereira
This article describes the first steps on the development of a low cost dataglove based on the Arduino Uno microprocessor. The glove is designed for use in virtual reality systems and is integrated into a suite of applications. This article also presents a set of preliminary results obtained with the glove and discuss about the use of the glove to control a video game application for rehabilitation of stroke patients.
{"title":"Development of a low cost dataglove based on arduino for virtual reality applications","authors":"Lucas Silva, R. Dantas, Andre L. H. Pantoja, Antonio Pereira","doi":"10.1109/CIVEMSA.2013.6617395","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617395","url":null,"abstract":"This article describes the first steps on the development of a low cost dataglove based on the Arduino Uno microprocessor. The glove is designed for use in virtual reality systems and is integrated into a suite of applications. This article also presents a set of preliminary results obtained with the glove and discuss about the use of the glove to control a video game application for rehabilitation of stroke patients.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132203327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617406
Yu Zhang, C. Bingham, M. Gallimore, Zhijing Yang, Jun Chen
The paper presents a readily implementable approach for sensor fault detection, identification (SFD/I) and faulted sensor data reconstruction in complex systems based on self-organizing map neural networks (SOMNNs). Two operational regimes are considered, i.e. the steady operation and operation with transients. For steady operation, SOMNN based estimation error (EE) are used for SFD. EE contribution plots are employed for SFI. For operation with transients, SOMNN classification maps are used for SFD/I comparing with the `fingerprint' maps. In addition, extension algorithm of SOMNNs is developed for faulted sensor data reconstruction. The validation of the proposed approach is demonstrated through experimental data during the commissioning of industrial gas turbines.
{"title":"Sensor fault detection and diagnosis based on SOMNNs for steady-state and transient operation","authors":"Yu Zhang, C. Bingham, M. Gallimore, Zhijing Yang, Jun Chen","doi":"10.1109/CIVEMSA.2013.6617406","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617406","url":null,"abstract":"The paper presents a readily implementable approach for sensor fault detection, identification (SFD/I) and faulted sensor data reconstruction in complex systems based on self-organizing map neural networks (SOMNNs). Two operational regimes are considered, i.e. the steady operation and operation with transients. For steady operation, SOMNN based estimation error (EE) are used for SFD. EE contribution plots are employed for SFI. For operation with transients, SOMNN classification maps are used for SFD/I comparing with the `fingerprint' maps. In addition, extension algorithm of SOMNNs is developed for faulted sensor data reconstruction. The validation of the proposed approach is demonstrated through experimental data during the commissioning of industrial gas turbines.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131943381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617419
Helal Saghir, D. Megherbi
Metagenomics is the study of microorganisms collected directly from natural environments. Metagenomics studies use DNA fragments obtained directly from a natural environment using whole genome shotgun (WGS) sequencing. Sequencing random fragments obtained from whole genome shotgun into taxa-based groups is known as binning. Currently, there are two different methods of binning: sequence similarity methods and sequence composition methods. Sequence similarity methods are usually based on sequence alignment to known genome like BLAST, or MEGAN. As only a very small fraction of species is available in the current databases, similarity methods do not yield good results. As a given database of organisms grows, the complexity of the search will also grow. Sequence composition methods are based on compositional features of a given DNA sequence like K-mers, or other genomic signature(s). Most of these current methods for binning have two major issues: they do not work well with short sequences and closely related genomes. In this paper we propose new machine learning related predictive DNA sequence feature selection algorithms to solve binning problems in more accurate and efficient ways. In this work we use Oligonucleotide frequencies from 2-mers to 4-mers as features to differentiate between sequences. 2-mers produces 16 features, 3-mers produces 64 features and 4-mers produces 256 features. We did not use feature higher than 4-mers as the number of feature increases exponentially and for 5-mers the number of feature would be 1024 features. We found out that the 4-mers produces better results than 2-mers and 3-mers. The data used in this work has an average length of 250, 500, 1000, and 2000 base pairs. Experimental results of the proposed algorithms are presented to show the potential value of the proposed methods. The proposed algorithm accuracy is tested on a variety of data sets and the classification/prediction accuracy achieved is between 78% - 99% for various simulated data sets using Random forest classifier and 37% - 95% using Naïve Bayes classifier. Random forest Classifier did better in classification in all the dataset compared to Naïve Bayes.
{"title":"An efficient comparative machine learning-based metagenomics binning technique via using Random forest","authors":"Helal Saghir, D. Megherbi","doi":"10.1109/CIVEMSA.2013.6617419","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617419","url":null,"abstract":"Metagenomics is the study of microorganisms collected directly from natural environments. Metagenomics studies use DNA fragments obtained directly from a natural environment using whole genome shotgun (WGS) sequencing. Sequencing random fragments obtained from whole genome shotgun into taxa-based groups is known as binning. Currently, there are two different methods of binning: sequence similarity methods and sequence composition methods. Sequence similarity methods are usually based on sequence alignment to known genome like BLAST, or MEGAN. As only a very small fraction of species is available in the current databases, similarity methods do not yield good results. As a given database of organisms grows, the complexity of the search will also grow. Sequence composition methods are based on compositional features of a given DNA sequence like K-mers, or other genomic signature(s). Most of these current methods for binning have two major issues: they do not work well with short sequences and closely related genomes. In this paper we propose new machine learning related predictive DNA sequence feature selection algorithms to solve binning problems in more accurate and efficient ways. In this work we use Oligonucleotide frequencies from 2-mers to 4-mers as features to differentiate between sequences. 2-mers produces 16 features, 3-mers produces 64 features and 4-mers produces 256 features. We did not use feature higher than 4-mers as the number of feature increases exponentially and for 5-mers the number of feature would be 1024 features. We found out that the 4-mers produces better results than 2-mers and 3-mers. The data used in this work has an average length of 250, 500, 1000, and 2000 base pairs. Experimental results of the proposed algorithms are presented to show the potential value of the proposed methods. The proposed algorithm accuracy is tested on a variety of data sets and the classification/prediction accuracy achieved is between 78% - 99% for various simulated data sets using Random forest classifier and 37% - 95% using Naïve Bayes classifier. Random forest Classifier did better in classification in all the dataset compared to Naïve Bayes.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132118238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2013-07-15DOI: 10.1109/CIVEMSA.2013.6617413
A. Stînean, S. Preitl, R. Precup, C. Dragos, M. Radac, E. Petriu
This paper treats the design and implementation of a low-cost neuro-fuzzy control solution for a class of servo systems with an integral component and variable parameters. A hybrid Takagi-Sugeno PI-neuro-fuzzy controller (T-S PI-N-FC) is proposed and presented along with its relatively simple design approach. The solution carries out the on-line adaptation of a single parameter of the input membership functions of a Takagi-Sugeno PI-fuzzy controller with input integration (T-S PI-FC-II) by a single neuron trained by back propagation with momentum factor in the framework of a model reference adaptive controller structure. The second parameter of the input membership functions is tuned by the modal equivalence principle. Linear matrix inequalities are proposed as sufficient stability conditions to be fulfilled by the parameters of the rule consequents of the T-S PI-FC-II in order to guarantee the stable design of the hybrid T-S PI-N-FC. The solution is validated by a case study using a set of three process parameters that correspond to a strip winding system laboratory equipment. Digital simulation results and experimental results are given.
本文研究一类具有整元变参数的伺服系统的低成本神经模糊控制方案的设计与实现。提出了一种混合Takagi-Sugeno pi -神经模糊控制器(T-S PI-N-FC)，并给出了其相对简单的设计方法。该方案实现了在模型参考自适应控制器结构框架内，通过带动量因子的反向传播训练的单个神经元在线自适应具有输入积分的Takagi-Sugeno pi -模糊控制器(T-S PI-FC-II)的单个参数的输入隶属度函数。输入隶属函数的第二个参数采用模态等效原理进行调谐。提出了线性矩阵不等式作为T-S PI-FC-II规则结果参数满足的充分稳定性条件，以保证混合T-S PI-N-FC的稳定设计。该解决方案通过一个案例研究进行了验证，该研究使用了一组三个工艺参数，这些参数对应于条带缠绕系统的实验室设备。给出了数字仿真结果和实验结果。
{"title":"Low-cost neuro-fuzzy control solution for servo systems with variable parameters","authors":"A. Stînean, S. Preitl, R. Precup, C. Dragos, M. Radac, E. Petriu","doi":"10.1109/CIVEMSA.2013.6617413","DOIUrl":"https://doi.org/10.1109/CIVEMSA.2013.6617413","url":null,"abstract":"This paper treats the design and implementation of a low-cost neuro-fuzzy control solution for a class of servo systems with an integral component and variable parameters. A hybrid Takagi-Sugeno PI-neuro-fuzzy controller (T-S PI-N-FC) is proposed and presented along with its relatively simple design approach. The solution carries out the on-line adaptation of a single parameter of the input membership functions of a Takagi-Sugeno PI-fuzzy controller with input integration (T-S PI-FC-II) by a single neuron trained by back propagation with momentum factor in the framework of a model reference adaptive controller structure. The second parameter of the input membership functions is tuned by the modal equivalence principle. Linear matrix inequalities are proposed as sufficient stability conditions to be fulfilled by the parameters of the rule consequents of the T-S PI-FC-II in order to guarantee the stable design of the hybrid T-S PI-N-FC. The solution is validated by a case study using a set of three process parameters that correspond to a strip winding system laboratory equipment. Digital simulation results and experimental results are given.","PeriodicalId":159100,"journal":{"name":"2013 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2013-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133823299","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：