Pub Date : 2017-05-01DOI: 10.1109/DDCLS.2017.8068133
W. Zhou, Miao Yu, Baobin Liu
In this paper, a direct learning control scheme is proposed for a class of discrete-time systems tracking magnitude-varying reference trajectories. The reference trajectories vary in magnitude with known ratios. A direct learning control algorithm is designed for systems with adequate pre-stored control information profiles, especially for systems working nonrepetitively in practice. By using pre-stored control input data, the new learning control input is developed directly. The simulation results of the permanent magnet linear motor show that the proposed direct learning control approach achieves perfect tracking.
{"title":"A discrete-time direct learning control scheme for magnitude-varying trajectories tracking","authors":"W. Zhou, Miao Yu, Baobin Liu","doi":"10.1109/DDCLS.2017.8068133","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068133","url":null,"abstract":"In this paper, a direct learning control scheme is proposed for a class of discrete-time systems tracking magnitude-varying reference trajectories. The reference trajectories vary in magnitude with known ratios. A direct learning control algorithm is designed for systems with adequate pre-stored control information profiles, especially for systems working nonrepetitively in practice. By using pre-stored control input data, the new learning control input is developed directly. The simulation results of the permanent magnet linear motor show that the proposed direct learning control approach achieves perfect tracking.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114603711","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068098
A. Zhang, Kailun Huang, Rui Wang, Zhiqiang Zhang
Due to the growth prospect of analog circuit fault diagnosis, this paper tends to introduce a novel arithmetic model based on least squares support vector machine (LSSVM) and the semi-supervised learning (SSL) scheme which is adept at cost-saving. The proposed method contains two steps. Firstly, the fact that large deviation may emerge as a result of the empirical risk inspires the idea of an improved transductive least square support vector machine (T-LSSVM) which aims at obtaining the best hyperplane that equipped with the maximum margin to the support vectors no matter whether the samples are labeled or unlabeled. Secondly, to overcome the drawback of typical T-LSSVM, i.e., sensitivity to local minima, a laplcian-transductive least squares support vector machine (Lap-T-LSSVM) is proposed which can perform the fault diagnosis via a laplcian. The experiment adopts band-pass filter circuit as diagnosis object. Simulation results verify that the proposed method is superior to previous SVM in accuracy.
{"title":"A novel hybrid method for analog circuit fault classification","authors":"A. Zhang, Kailun Huang, Rui Wang, Zhiqiang Zhang","doi":"10.1109/DDCLS.2017.8068098","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068098","url":null,"abstract":"Due to the growth prospect of analog circuit fault diagnosis, this paper tends to introduce a novel arithmetic model based on least squares support vector machine (LSSVM) and the semi-supervised learning (SSL) scheme which is adept at cost-saving. The proposed method contains two steps. Firstly, the fact that large deviation may emerge as a result of the empirical risk inspires the idea of an improved transductive least square support vector machine (T-LSSVM) which aims at obtaining the best hyperplane that equipped with the maximum margin to the support vectors no matter whether the samples are labeled or unlabeled. Secondly, to overcome the drawback of typical T-LSSVM, i.e., sensitivity to local minima, a laplcian-transductive least squares support vector machine (Lap-T-LSSVM) is proposed which can perform the fault diagnosis via a laplcian. The experiment adopts band-pass filter circuit as diagnosis object. Simulation results verify that the proposed method is superior to previous SVM in accuracy.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131100960","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068167
Li Wang, Ke Pan, Xingyu Wang
With the development of satellite positioning technology and network communication technology, the floating vehicle data has the advantages of large amount of data, high precision and wide Coverage. In order to improve the green wave control effect, combines the floating vehicle trajectory data with distributed wave theory, makes the real-time prediction of queue length of current cycle and the time of complete disappearance, and analyzes the green-start coordination, green-middle coordination and green-end coordination in the advantages and disadvantages, designs an optimal control strategy with dynamic green point floating vehicle queue length based on real-time sensing green, improves the efficiency of green band, through simulation experiments to prove the effectiveness of the proposed method.
{"title":"Real-time queue length perception with green wave band point optimization based on floating vehicle","authors":"Li Wang, Ke Pan, Xingyu Wang","doi":"10.1109/DDCLS.2017.8068167","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068167","url":null,"abstract":"With the development of satellite positioning technology and network communication technology, the floating vehicle data has the advantages of large amount of data, high precision and wide Coverage. In order to improve the green wave control effect, combines the floating vehicle trajectory data with distributed wave theory, makes the real-time prediction of queue length of current cycle and the time of complete disappearance, and analyzes the green-start coordination, green-middle coordination and green-end coordination in the advantages and disadvantages, designs an optimal control strategy with dynamic green point floating vehicle queue length based on real-time sensing green, improves the efficiency of green band, through simulation experiments to prove the effectiveness of the proposed method.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128427360","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068158
Junyuan Tang, Jun Wu, Shengjun Huang
Maglev train is a new urban transportation tool. Track is considered as a component of maglev transportation system and it would directly affect the safety of train operation. An algorithm based on triple threshold filtering is put forward to conduct detection of maglev track, which can be used to filter abnormal points to judge the vertical suspected irregularity of track. In the scheme, the threshold settings for gap difference, current change rate and acceleration differences on suspended controller are set according to characteristics of low sampling rate, high data repeatability and large data volume of automobile data recorder, it can extract information of track irregularity. In order to improve the reliability of the algorithm, 20 sets of data from 5 independent bogies of two trains are clustered for analysis, and get the suspected area of vertical irregularity. The data of maglev train of Changsha operating railway was tested finally.
{"title":"On middle and low speed maglev track irregularity detection based on threshold filtering algorithm","authors":"Junyuan Tang, Jun Wu, Shengjun Huang","doi":"10.1109/DDCLS.2017.8068158","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068158","url":null,"abstract":"Maglev train is a new urban transportation tool. Track is considered as a component of maglev transportation system and it would directly affect the safety of train operation. An algorithm based on triple threshold filtering is put forward to conduct detection of maglev track, which can be used to filter abnormal points to judge the vertical suspected irregularity of track. In the scheme, the threshold settings for gap difference, current change rate and acceleration differences on suspended controller are set according to characteristics of low sampling rate, high data repeatability and large data volume of automobile data recorder, it can extract information of track irregularity. In order to improve the reliability of the algorithm, 20 sets of data from 5 independent bogies of two trains are clustered for analysis, and get the suspected area of vertical irregularity. The data of maglev train of Changsha operating railway was tested finally.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126392814","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068128
Chunlin Du, Ying Nan, Rui Yan
This paper proposes a biologically plausible network architecture with spiking neurons for face recognition. This network consists of three parts: feature extraction, encoding and classification. Firstly, HMAX model with four layers (C1-S1-C2-S2) is used to extract face features. The proposed feature extraction method can keep selectivity invariance and scale invariance. The next important part is to encode features to suitable spike trains for spiking neural networks. In the last part, the improved Tempotron learning rule is chosen to train the spiking neural networks with reduced computation and increased fault tolerance. In order to demonstrate the performance of spiking neural networks, four databases are tested in the experiment: Yale, Extend Yale B, ORL, and FERET.
{"title":"Spike-based learning rules for face recognition","authors":"Chunlin Du, Ying Nan, Rui Yan","doi":"10.1109/DDCLS.2017.8068128","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068128","url":null,"abstract":"This paper proposes a biologically plausible network architecture with spiking neurons for face recognition. This network consists of three parts: feature extraction, encoding and classification. Firstly, HMAX model with four layers (C1-S1-C2-S2) is used to extract face features. The proposed feature extraction method can keep selectivity invariance and scale invariance. The next important part is to encode features to suitable spike trains for spiking neural networks. In the last part, the improved Tempotron learning rule is chosen to train the spiking neural networks with reduced computation and increased fault tolerance. In order to demonstrate the performance of spiking neural networks, four databases are tested in the experiment: Yale, Extend Yale B, ORL, and FERET.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126636623","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068140
Darong Huang, Dongjie Zhao, Ling Zhao
In this paper, the optimal route and deployment scheme are designed to ensure the shortest retention time for unmanned aerial vehicles (UAV) in risk area. Firstly, according to the known data and radar scanning range, the regional distribution map of target grope and base are obtained, respectively. Secondly, based on the different scanning bandwidth of loads, target points are classified by using clustering analysis. This makes the target points fall on the scanning bandwidth of UAV as far as possible, accordingly reducing the UAV's scanning times. This problem can be regarded as a travelling salesman problem in radar scanning range. Finally, the deployment result and locally optimal route are obtained by 0–1 programming in LINGO. Furthermore, particle swarm optimization is used to improve the local optimal path and the global optimal route can then be generated.
{"title":"A new method of the shortest path planning for unmanned aerial vehicles","authors":"Darong Huang, Dongjie Zhao, Ling Zhao","doi":"10.1109/DDCLS.2017.8068140","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068140","url":null,"abstract":"In this paper, the optimal route and deployment scheme are designed to ensure the shortest retention time for unmanned aerial vehicles (UAV) in risk area. Firstly, according to the known data and radar scanning range, the regional distribution map of target grope and base are obtained, respectively. Secondly, based on the different scanning bandwidth of loads, target points are classified by using clustering analysis. This makes the target points fall on the scanning bandwidth of UAV as far as possible, accordingly reducing the UAV's scanning times. This problem can be regarded as a travelling salesman problem in radar scanning range. Finally, the deployment result and locally optimal route are obtained by 0–1 programming in LINGO. Furthermore, particle swarm optimization is used to improve the local optimal path and the global optimal route can then be generated.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126880182","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068118
Haiyan Wu, Yu Chen, Jing Wang
The feedback control of molecular weight distribution (MWD) of polymerization is considered in this paper. Based on the neural network modeling, the control of MWD can be realized by the output feedback control (OFC) with 2–3 moments as the control object. However, the control system, designed from the neural network model, has a low reliability and accuracy owing to the modeling errors and unmodeled dynamics. As such, a practical modularized control scheme is demonstrate by adding the model-free control (MFC) into the existing OFC schemes for enjoying the extra performance improvement from MFC. The new control method is tested on styrene polymerization reacted in CSTR, and simulation results proved the effectiveness of the method. Furthermore, the proposed control method has an enhanced reliability and precision compared with OFC method.
{"title":"Model-free output feedback control of molecular weight distribution","authors":"Haiyan Wu, Yu Chen, Jing Wang","doi":"10.1109/DDCLS.2017.8068118","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068118","url":null,"abstract":"The feedback control of molecular weight distribution (MWD) of polymerization is considered in this paper. Based on the neural network modeling, the control of MWD can be realized by the output feedback control (OFC) with 2–3 moments as the control object. However, the control system, designed from the neural network model, has a low reliability and accuracy owing to the modeling errors and unmodeled dynamics. As such, a practical modularized control scheme is demonstrate by adding the model-free control (MFC) into the existing OFC schemes for enjoying the extra performance improvement from MFC. The new control method is tested on styrene polymerization reacted in CSTR, and simulation results proved the effectiveness of the method. Furthermore, the proposed control method has an enhanced reliability and precision compared with OFC method.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"79 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121341200","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068125
Jie Yang, Zhihuan Song, Li Jiang
Restrict Boltzmann Machine, a generative model that consists of one visible layer and one hidden layer, plays an important role in deep learning. It can be used as a feature extractor in an unsupervised way. In process diagnosis area, the Sparse Class Gaussian Restrict Boltzmann Machine is developed as a discriminative nonlinear feature extractor for classification in order to solve the discriminative task. Moreover, for the purpose of overcoming the overfitting and raising the training efficiency, a sparse constraint for the hidden layer is added during the training time. The experimental results based on TE process benchmark successfully demonstrate that this model significantly outperforms the MLP classifier, and other GRBM based models, moreover, the sparse constraint employed has a positive effect on the classification performance.
{"title":"Fault diagnosis based on sparse class Gaussian Restrict Boltzmann Machine model","authors":"Jie Yang, Zhihuan Song, Li Jiang","doi":"10.1109/DDCLS.2017.8068125","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068125","url":null,"abstract":"Restrict Boltzmann Machine, a generative model that consists of one visible layer and one hidden layer, plays an important role in deep learning. It can be used as a feature extractor in an unsupervised way. In process diagnosis area, the Sparse Class Gaussian Restrict Boltzmann Machine is developed as a discriminative nonlinear feature extractor for classification in order to solve the discriminative task. Moreover, for the purpose of overcoming the overfitting and raising the training efficiency, a sparse constraint for the hidden layer is added during the training time. The experimental results based on TE process benchmark successfully demonstrate that this model significantly outperforms the MLP classifier, and other GRBM based models, moreover, the sparse constraint employed has a positive effect on the classification performance.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128941673","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068138
Yang Zhao, F. Zhou, Da Wang, Yan Li
In this paper, we present an iterative learning control (ILC) frameworks for robust path-tracking problem of nonholonomic mobile robots in the presence of initial shifts. The major difficulties are owning to the simultaneous existence of state disturbances and biased initial state in the mobile robot kinematic system. To design ILC strategy for such problem, a new ILC scheme is proposed as a combination of an initial rectifying term and a feedback-aided P-type learning algorithm. Sufficient conditions of convergence of this approach are given and the global convergence is proved. Simulation results also verify the effectiveness of the proposed scheme.
{"title":"Path-tracking of mobile robot using feedback-aided P-type iterative learning control against initial state error","authors":"Yang Zhao, F. Zhou, Da Wang, Yan Li","doi":"10.1109/DDCLS.2017.8068138","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068138","url":null,"abstract":"In this paper, we present an iterative learning control (ILC) frameworks for robust path-tracking problem of nonholonomic mobile robots in the presence of initial shifts. The major difficulties are owning to the simultaneous existence of state disturbances and biased initial state in the mobile robot kinematic system. To design ILC strategy for such problem, a new ILC scheme is proposed as a combination of an initial rectifying term and a feedback-aided P-type learning algorithm. Sufficient conditions of convergence of this approach are given and the global convergence is proved. Simulation results also verify the effectiveness of the proposed scheme.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114798057","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068064
L. Zhang, M. Li
This paper is concerned with the problem of fuzzy modeling in the T-S fuzzy systems. In order to obtain relative simple T-S fuzzy model, combined with the traditional linearization approximation method on the fuzzy modeling. It deals the nonlinear term of unable to approximate with the same method of the uncertainties of processing, so that it could design the uncertainty T-S fuzzy system with the robust technology. Although the presented modeling method will obtain larger dimension matrix inequality, it can be decreased the number of fuzzy rules. The matrix inequality dimension and the number of fuzzy rules would be discussed in a simulation example based on the comparison of control calculation burden.
{"title":"Fuzzy modeling and control of a class of simple pendulum system based on robust technology","authors":"L. Zhang, M. Li","doi":"10.1109/DDCLS.2017.8068064","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068064","url":null,"abstract":"This paper is concerned with the problem of fuzzy modeling in the T-S fuzzy systems. In order to obtain relative simple T-S fuzzy model, combined with the traditional linearization approximation method on the fuzzy modeling. It deals the nonlinear term of unable to approximate with the same method of the uncertainties of processing, so that it could design the uncertainty T-S fuzzy system with the robust technology. Although the presented modeling method will obtain larger dimension matrix inequality, it can be decreased the number of fuzzy rules. The matrix inequality dimension and the number of fuzzy rules would be discussed in a simulation example based on the comparison of control calculation burden.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126880361","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：