Pub Date : 2006-10-23DOI: 10.1109/WCICA.2006.1713848
Huajie Chen, Wei Wei
In order to increase the detection accuracy in gender classification, a pseudo-example based iterative learning approach combining support vector machine (SVM) and active appearance model (AAM) was proposed. AAM was applied to model the original training examples before constructing the SVM classifier. During the current iteration, some pairs of support vectors with different gender were selected randomly and then their AAM parameters were interpolated properly to generate new pseudo face images as candidate examples with new gender feature pattern. Only the candidates that would be classified by the current classifier incorrectly or correctly but with low confidence were selected for the following iterations. The pseudo-examples created in this way complemented the original training examples effectively, and the proposed pseudo-example selecting scheme outperformed the conventional Bootstrap method. Experimental results show that, this iterative learning approach can upgrade the gender detection accuracy stepwise
{"title":"Pseudo-Example Based Iterative SVM Learning Approach for Gender Classification","authors":"Huajie Chen, Wei Wei","doi":"10.1109/WCICA.2006.1713848","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713848","url":null,"abstract":"In order to increase the detection accuracy in gender classification, a pseudo-example based iterative learning approach combining support vector machine (SVM) and active appearance model (AAM) was proposed. AAM was applied to model the original training examples before constructing the SVM classifier. During the current iteration, some pairs of support vectors with different gender were selected randomly and then their AAM parameters were interpolated properly to generate new pseudo face images as candidate examples with new gender feature pattern. Only the candidates that would be classified by the current classifier incorrectly or correctly but with low confidence were selected for the following iterations. The pseudo-examples created in this way complemented the original training examples effectively, and the proposed pseudo-example selecting scheme outperformed the conventional Bootstrap method. Experimental results show that, this iterative learning approach can upgrade the gender detection accuracy stepwise","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129650865","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1713801
Xingli Huang, Jihong Zhu, Shiqian Liu, P. Jia
Based on a small unmanned helicopter hovering ground testbed, considering strong dynamic couplings between rotors and body, the front rotor and the rear rotor of the small unmanned tandem helicopter, a nonlinear dynamic model of hovering small unmanned rotor helicopter was built by Newton law and Lagrange algorithm. A dynamic inversion method was employed to design the corresponding nonlinear flight control law. And a RBF neural network with on-line learning capability was designed to overcome the influences of exterior disturbance and uncertainty of modeling. Simulation results demonstrate that the instruction tracking behaviors are improved under constraints of desired requirements and the obtained results are verified
{"title":"Neural Network Adaptive Control for Small Unmanned Tandem Helicopter","authors":"Xingli Huang, Jihong Zhu, Shiqian Liu, P. Jia","doi":"10.1109/WCICA.2006.1713801","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713801","url":null,"abstract":"Based on a small unmanned helicopter hovering ground testbed, considering strong dynamic couplings between rotors and body, the front rotor and the rear rotor of the small unmanned tandem helicopter, a nonlinear dynamic model of hovering small unmanned rotor helicopter was built by Newton law and Lagrange algorithm. A dynamic inversion method was employed to design the corresponding nonlinear flight control law. And a RBF neural network with on-line learning capability was designed to overcome the influences of exterior disturbance and uncertainty of modeling. Simulation results demonstrate that the instruction tracking behaviors are improved under constraints of desired requirements and the obtained results are verified","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129884470","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1713567
Ruihua Li, Guoxiang Meng, Zhengjin Feng, Yijie Li, Weixiang Shi
A sliding mode control (SMC) approach was used in a pneumatic force control servo-system with uncertainties and external disturbances in this paper. To reach exact force control in pneumatic force servo-system, the E/P pressure proportional valve was used as an electro-pneumatic converter in constructing an electro-pneumatic force control servo system. The dynamic model of the pneumatic force servo system was built firstly, and then a boundary layer approach was used to the design of controller, in which using a saturation function to substitute for the sign function based on Lyapunov theory. Finally, the performance of the controller was simulated and tested. The results show that the system performance has been much improved when compared with a conventional PID controller, and it has a good control precision and robustness
{"title":"A Sliding Mode Variable Structure Control Approach for a Pneumatic Force Servo System","authors":"Ruihua Li, Guoxiang Meng, Zhengjin Feng, Yijie Li, Weixiang Shi","doi":"10.1109/WCICA.2006.1713567","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713567","url":null,"abstract":"A sliding mode control (SMC) approach was used in a pneumatic force control servo-system with uncertainties and external disturbances in this paper. To reach exact force control in pneumatic force servo-system, the E/P pressure proportional valve was used as an electro-pneumatic converter in constructing an electro-pneumatic force control servo system. The dynamic model of the pneumatic force servo system was built firstly, and then a boundary layer approach was used to the design of controller, in which using a saturation function to substitute for the sign function based on Lyapunov theory. Finally, the performance of the controller was simulated and tested. The results show that the system performance has been much improved when compared with a conventional PID controller, and it has a good control precision and robustness","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129920120","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1714270
Xiuhua Wu, Guokai Sun, Z. Piao, Zhaoyuan Liu, Ping Yang
Using the BML (Biham, Middleton and Levine) model, a special one of the cellular automata models, the paper simulates and analyzes the two-dimensional traffic system controlled by traffic lights. The relationship among original density, average density and average speed of the traffic flow can be found through computer programs. It can be seen that self-organization phenomenon existed in the processing from the traffic jam status to the movement status when the average density is kept invariable. Then, it analyzes the relationship between the position of critical point and the average density. Finally, it shows the function of critical density in real traffic control
利用元胞自动机模型中的一种特殊模型BML (Biham, Middleton and Levine)模型，对二维交通信号灯控制的交通系统进行了仿真分析。通过计算机程序可以求出交通流的原始密度、平均密度和平均速度之间的关系。可以看出，在保持平均密度不变的情况下，从拥堵状态到运动状态的处理过程中存在自组织现象。然后，分析了临界点位置与平均密度的关系。最后，给出了临界密度在实际交通控制中的作用
{"title":"The Application of the Cellular Automata in Simulating and Analyzing Two-Dimensional Traffic Model","authors":"Xiuhua Wu, Guokai Sun, Z. Piao, Zhaoyuan Liu, Ping Yang","doi":"10.1109/WCICA.2006.1714270","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1714270","url":null,"abstract":"Using the BML (Biham, Middleton and Levine) model, a special one of the cellular automata models, the paper simulates and analyzes the two-dimensional traffic system controlled by traffic lights. The relationship among original density, average density and average speed of the traffic flow can be found through computer programs. It can be seen that self-organization phenomenon existed in the processing from the traffic jam status to the movement status when the average density is kept invariable. Then, it analyzes the relationship between the position of critical point and the average density. Finally, it shows the function of critical density in real traffic control","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130421991","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1712350
Yong Liu, Rui Huang, Jim Zhu
In this paper, an adaptive neural network nonlinear control method is developed based on trajectory linearization control (TLC). The adaptive neural network TLC control (ANNTLC) compensates the model nonlinear uncertainty adaptively, and improves controller performance. ANNTLC can also be used to simplify the TLC control design procedure by using a simplified model. A stable neural network learning rule is developed. The simulation result shows the feasibility of the proposed method
{"title":"Adaptive Neural Network Control Based on Trajectory Linearization Control","authors":"Yong Liu, Rui Huang, Jim Zhu","doi":"10.1109/WCICA.2006.1712350","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1712350","url":null,"abstract":"In this paper, an adaptive neural network nonlinear control method is developed based on trajectory linearization control (TLC). The adaptive neural network TLC control (ANNTLC) compensates the model nonlinear uncertainty adaptively, and improves controller performance. ANNTLC can also be used to simplify the TLC control design procedure by using a simplified model. A stable neural network learning rule is developed. The simulation result shows the feasibility of the proposed method","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130512371","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1714229
Bin Sun, Yuxiao Zhao
For identifying gas-liquid two-phase flow regime, a kind of data mining method based on Hilbert-Huang transform was put forward. At first, the dynamic differential pressure signal coming from Venturitube was handled by HHT, the instantaneous frequency, instantaneous swing and user-defined characteristic variable of different mode were calculated. The relation of characteristic variable and flow regime was obtained by visual analyzing instantaneous frequency and characteristic variable. Afterwards, according to the distribution of characteristic variable, fuzzy association rules of identifying flow regime was gained adopting data mining method, and the results of flow regime identification were acquire through fuzzy illation and calculation. The experimental results of gas-water two-phase flow in vertical pipes with 50mm and 40mm inner diameter show, this method could identify bubble flow, slug flow and churn flow effectively, and discriminating precision exceed 94%. The method's principle is easy, has few influence by experimental condition and good universality, and it could settle for practical flow regime identification
{"title":"Data Mining Method Based on HHT and Application Research in Flow Regime Identification","authors":"Bin Sun, Yuxiao Zhao","doi":"10.1109/WCICA.2006.1714229","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1714229","url":null,"abstract":"For identifying gas-liquid two-phase flow regime, a kind of data mining method based on Hilbert-Huang transform was put forward. At first, the dynamic differential pressure signal coming from Venturitube was handled by HHT, the instantaneous frequency, instantaneous swing and user-defined characteristic variable of different mode were calculated. The relation of characteristic variable and flow regime was obtained by visual analyzing instantaneous frequency and characteristic variable. Afterwards, according to the distribution of characteristic variable, fuzzy association rules of identifying flow regime was gained adopting data mining method, and the results of flow regime identification were acquire through fuzzy illation and calculation. The experimental results of gas-water two-phase flow in vertical pipes with 50mm and 40mm inner diameter show, this method could identify bubble flow, slug flow and churn flow effectively, and discriminating precision exceed 94%. The method's principle is easy, has few influence by experimental condition and good universality, and it could settle for practical flow regime identification","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126694434","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1713542
Qiang Liu, P. Feng, S. Pan
When implementing different tasks, mechanical servo systems must adapt to various working loads with different weight or inertia, which may lead to the remarkable varying of inertial parameters. However, for such cases, motion control methods at present such as PD control and disturbance observer based robust control design, may exhibit instability or decline of tracking performance. For the problems mentioned above, a novel nonlinear control scheme, for which the varying range of inertial parameters was supposed to known, was presented. The nonlinear controller is composed of two parts: the PD control design for the reference model system, and the sliding mode control of the mechanical plant. The sliding mode technique was used for servo system to achieve robust stability and guaranteed transient response, and the boundary layer control was adopted to avoid chattering introduced by control switching. The global stability of the system is proved, and the transient performance is analyzed. Computer simulation results developed for a DC motor servo system show the effectiveness of the proposed method
{"title":"Robust Motion Control of High Precision Mechanical Servo Systems with Parameter Uncertainties","authors":"Qiang Liu, P. Feng, S. Pan","doi":"10.1109/WCICA.2006.1713542","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713542","url":null,"abstract":"When implementing different tasks, mechanical servo systems must adapt to various working loads with different weight or inertia, which may lead to the remarkable varying of inertial parameters. However, for such cases, motion control methods at present such as PD control and disturbance observer based robust control design, may exhibit instability or decline of tracking performance. For the problems mentioned above, a novel nonlinear control scheme, for which the varying range of inertial parameters was supposed to known, was presented. The nonlinear controller is composed of two parts: the PD control design for the reference model system, and the sliding mode control of the mechanical plant. The sliding mode technique was used for servo system to achieve robust stability and guaranteed transient response, and the boundary layer control was adopted to avoid chattering introduced by control switching. The global stability of the system is proved, and the transient performance is analyzed. Computer simulation results developed for a DC motor servo system show the effectiveness of the proposed method","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126797560","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1713899
Fan Zhang, Jianjun Zhao, Xinhong Zhang
A blind digital watermarking algorithm is presented based on the perceptual models and Hopfield neural network. The Hopfield neural network stores the host image and the original watermark. The noise visibility function (NVF) is used for adaptively watermark embedding. In the watermark extraction, the host image and the original watermark are retrieved by neural network. The experimental results show that this watermarking algorithm has a good preferment
{"title":"Adaptive Watermarking Algorithm Based on Perceptual Models","authors":"Fan Zhang, Jianjun Zhao, Xinhong Zhang","doi":"10.1109/WCICA.2006.1713899","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713899","url":null,"abstract":"A blind digital watermarking algorithm is presented based on the perceptual models and Hopfield neural network. The Hopfield neural network stores the host image and the original watermark. The noise visibility function (NVF) is used for adaptively watermark embedding. In the watermark extraction, the host image and the original watermark are retrieved by neural network. The experimental results show that this watermarking algorithm has a good preferment","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126992575","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1713754
Hongli Zhou, Ge Guo, Manqiang Liu
Multi-robot systems differ from single robot systems mostly in that the environments can be affected by other robots. So we can consider every robot in dynamic environments. Therefore it is crucial that each robot should have both learning and evolutionary ability to adapt to dynamic environments. This paper proposes a new robot behavior decision controller using Elman neural network (Elman NN) and genetic algorithm (GA).The Elman NN has the advantages of time series prediction capability because of its memory nodes, as well as local recurrent connections. Genetic algorithm (GA) is introduced to determine the connection weight values of Elman NN in order to achieve better behavior performance. The computer simulation is given to show the validity of the method
{"title":"GA-Aided Elman Neural Network Controller For Behavior-Based Robot","authors":"Hongli Zhou, Ge Guo, Manqiang Liu","doi":"10.1109/WCICA.2006.1713754","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1713754","url":null,"abstract":"Multi-robot systems differ from single robot systems mostly in that the environments can be affected by other robots. So we can consider every robot in dynamic environments. Therefore it is crucial that each robot should have both learning and evolutionary ability to adapt to dynamic environments. This paper proposes a new robot behavior decision controller using Elman neural network (Elman NN) and genetic algorithm (GA).The Elman NN has the advantages of time series prediction capability because of its memory nodes, as well as local recurrent connections. Genetic algorithm (GA) is introduced to determine the connection weight values of Elman NN in order to achieve better behavior performance. The computer simulation is given to show the validity of the method","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126993686","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 : 2006-10-23DOI: 10.1109/WCICA.2006.1712877
Q. Zong, Wenjing Liu, Liqian Dou
Support vector machine (SVM) has recently emerged as a powerful technique for solving problems in pattern classification and regression, but its performance mainly depends on the parameters selection of it. Parameters selection for SVM is very complex in nature and quite hard to solve by conventional optimization techniques, which constrains its application to some degree. PSO, as an evolutionary computing technology, has been applied successfully to various optimization problems, but has some disadvantage. So in this paper PSO is modified by added certain particles at each iterative to broaden search area, which makes particles free of local optimization. A new methodology for parameters selection of support vector regression is proposed, based on the modified PSO tuning algorithm. The methodology is used to model nonlinear dynamical system in simulation, and the simulation result assures the validity of it, not only on time but also on model accuracy
{"title":"Parameters selection for SVR based on PSO","authors":"Q. Zong, Wenjing Liu, Liqian Dou","doi":"10.1109/WCICA.2006.1712877","DOIUrl":"https://doi.org/10.1109/WCICA.2006.1712877","url":null,"abstract":"Support vector machine (SVM) has recently emerged as a powerful technique for solving problems in pattern classification and regression, but its performance mainly depends on the parameters selection of it. Parameters selection for SVM is very complex in nature and quite hard to solve by conventional optimization techniques, which constrains its application to some degree. PSO, as an evolutionary computing technology, has been applied successfully to various optimization problems, but has some disadvantage. So in this paper PSO is modified by added certain particles at each iterative to broaden search area, which makes particles free of local optimization. A new methodology for parameters selection of support vector regression is proposed, based on the modified PSO tuning algorithm. The methodology is used to model nonlinear dynamical system in simulation, and the simulation result assures the validity of it, not only on time but also on model accuracy","PeriodicalId":375135,"journal":{"name":"2006 6th World Congress on Intelligent Control and Automation","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-10-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127040947","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：