We consider estimating a stochastic intensity process, indirectly observed through a sequence of Poisson random variables. The class of dynamical systems we consider, are autoregressive state processes and a sequence of Poisson distributed observations, each influenced by the state process. Computing explicit recursive filters for this class of system can be technically difficult. Further, exact filters for this class of model do not have fixed memory requirements. It is shown in Manton et al. (1999), that the exact filter for these models has linear growth with respect to time in the number of sufficient statistics needed to construct the filter density. Some approximate suboptimal filters are available for these models using Edgeworth expansions. Using Sequential Monte Carlo (SMC) methods and the so called Bayesian bootstrap filter, we propose a scheme to estimate an intensity process for the models just described. The scheme we present is simple to implement and has fixed memory requirements. A simulation study is included to demonstrate the performance of SMC methods for discrete time models with Poisson observations.
{"title":"Sequential Monte Carlo state estimation for Poisson observations at discrete times","authors":"W. P. Malcolm, N. Gordon","doi":"10.1109/IDC.2002.995414","DOIUrl":"https://doi.org/10.1109/IDC.2002.995414","url":null,"abstract":"We consider estimating a stochastic intensity process, indirectly observed through a sequence of Poisson random variables. The class of dynamical systems we consider, are autoregressive state processes and a sequence of Poisson distributed observations, each influenced by the state process. Computing explicit recursive filters for this class of system can be technically difficult. Further, exact filters for this class of model do not have fixed memory requirements. It is shown in Manton et al. (1999), that the exact filter for these models has linear growth with respect to time in the number of sufficient statistics needed to construct the filter density. Some approximate suboptimal filters are available for these models using Edgeworth expansions. Using Sequential Monte Carlo (SMC) methods and the so called Bayesian bootstrap filter, we propose a scheme to estimate an intensity process for the models just described. The scheme we present is simple to implement and has fixed memory requirements. A simulation study is included to demonstrate the performance of SMC methods for discrete time models with Poisson observations.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126856297","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}
The paper considers the controller design process in the problem of robust active noise control in an acoustic duct. In particular, the paper looks at how differences in system identification and actuator location affects the disturbance attenuation performance of the resulting control system. The robust controller design methodology being considered is based on some results on minimax LQG control.
{"title":"System identification, uncertainty modelling and actuator placement in the robust control of an acoustic duct","authors":"I. Petersen, H. Pota, M. Jahromi","doi":"10.1109/IDC.2002.995363","DOIUrl":"https://doi.org/10.1109/IDC.2002.995363","url":null,"abstract":"The paper considers the controller design process in the problem of robust active noise control in an acoustic duct. In particular, the paper looks at how differences in system identification and actuator location affects the disturbance attenuation performance of the resulting control system. The robust controller design methodology being considered is based on some results on minimax LQG control.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125740298","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}
The work leads with three dimensional environments for industrial processes visualization and the study of control parameters for the optimisation of real time systems simulation with evolutionary approach. The initial motivation behind the development of adaptive control models was the need to account for uncertainty in the parameters and structure of physical systems. When such a system to be controlled should be simulated by a computer with some stochastic model, then it is suitable to use evolutionary computing to determine automatically the role of each parameter on the system performance, what can lead to an advantage in the optimisation of some proposed automation model under study. Concurrently the use of virtual environments with the visualization of the model's performance in a three dimensional perspective could open the possibility for the user to taste some kind of immersion as close as possible into the reality of the system in simulation.
{"title":"A virtual environment for evolutionary autonomous optimization of real time stochastic control design","authors":"A.C.D. Silva","doi":"10.1109/IDC.2002.995370","DOIUrl":"https://doi.org/10.1109/IDC.2002.995370","url":null,"abstract":"The work leads with three dimensional environments for industrial processes visualization and the study of control parameters for the optimisation of real time systems simulation with evolutionary approach. The initial motivation behind the development of adaptive control models was the need to account for uncertainty in the parameters and structure of physical systems. When such a system to be controlled should be simulated by a computer with some stochastic model, then it is suitable to use evolutionary computing to determine automatically the role of each parameter on the system performance, what can lead to an advantage in the optimisation of some proposed automation model under study. Concurrently the use of virtual environments with the visualization of the model's performance in a three dimensional perspective could open the possibility for the user to taste some kind of immersion as close as possible into the reality of the system in simulation.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125473522","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}
Considers controllability properties of linear impulsive dynamical systems. As the controllability of linear impulsive systems is trivial, the investigation is focused on the controllability properties when impulsive vectors are applied in the sub systems of the linear impulsive systems. The controllability conditions including when there is a feedback in the system are derived.
{"title":"Controllability of linear impulsive systems","authors":"E. Joelianto, D. Williamson","doi":"10.1109/IDC.2002.995397","DOIUrl":"https://doi.org/10.1109/IDC.2002.995397","url":null,"abstract":"Considers controllability properties of linear impulsive dynamical systems. As the controllability of linear impulsive systems is trivial, the investigation is focused on the controllability properties when impulsive vectors are applied in the sub systems of the linear impulsive systems. The controllability conditions including when there is a feedback in the system are derived.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"119 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123471252","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}
A problem of task rescheduling using a coordinator in a structural decentralised control of discrete-event systems is formulated. After the local supervisors have been designed for a given task, a coordinator can be used to solve some rescheduling problems among local plants. The coordination system models the interactions of local plants, and is consisting of only the shared events, so simpler to synthesise. Under the conditions developed in this paper, the combined concurrent actions of the coordinator with the existing local supervisors can achieve the rescheduling requirements. An example is provided for illustration.
{"title":"Task rescheduling using a coordinator in a structural decentralised control of discrete-event systems","authors":"Sang-Heon Lee, Ill-Soo Kim, K.C. Wong","doi":"10.1109/IDC.2002.995401","DOIUrl":"https://doi.org/10.1109/IDC.2002.995401","url":null,"abstract":"A problem of task rescheduling using a coordinator in a structural decentralised control of discrete-event systems is formulated. After the local supervisors have been designed for a given task, a coordinator can be used to solve some rescheduling problems among local plants. The coordination system models the interactions of local plants, and is consisting of only the shared events, so simpler to synthesise. Under the conditions developed in this paper, the combined concurrent actions of the coordinator with the existing local supervisors can achieve the rescheduling requirements. An example is provided for illustration.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124651767","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}
Power gain is a simple generalization of L/sub 2/-gain that admits the analysis of a class of nonlinear systems with dynamics that exhibit an attracting set. In the paper, a notion of power gain is defined for interconnected systems. This power gain inequality allows the behaviour of the interconnected system to be described in terms of the component subsystems. A small gain theorem follows and an example is presented.
{"title":"A remark on a power gain inequality for interconnected systems","authors":"P. Dower","doi":"10.1109/IDC.2002.995383","DOIUrl":"https://doi.org/10.1109/IDC.2002.995383","url":null,"abstract":"Power gain is a simple generalization of L/sub 2/-gain that admits the analysis of a class of nonlinear systems with dynamics that exhibit an attracting set. In the paper, a notion of power gain is defined for interconnected systems. This power gain inequality allows the behaviour of the interconnected system to be described in terms of the component subsystems. A small gain theorem follows and an example is presented.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121806011","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}
Introduces an application of fuzzy control ideas to the precision missile guidance problem with a maneuvering target. The simple fuzzy controller uses angle only measurements in the control strategy and produced satisfactory results in comparison to the LQR or H/sup /spl infin// type precision guidance controllers.
{"title":"Fuzzy precision guidance controller with angle only measurements","authors":"P. N. Pathirana, A. Savkin","doi":"10.1109/IDC.2002.995409","DOIUrl":"https://doi.org/10.1109/IDC.2002.995409","url":null,"abstract":"Introduces an application of fuzzy control ideas to the precision missile guidance problem with a maneuvering target. The simple fuzzy controller uses angle only measurements in the control strategy and produced satisfactory results in comparison to the LQR or H/sup /spl infin// type precision guidance controllers.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"403 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125317873","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}
Statistical mixture-of-experts models are often used for data analysis tasks such as clustering, regression and classification. We consider two mixture-of-experts models, the shared mixture classifier and the hierarchical mixture-of-experts classifier. We discuss the initialisation and optimisation of the structure and parameters of each classifier. In particular, we initialise the hierarchical mixture of experts classifier with the public domain OC1 decision tree software. We compare the performance of the two classifiers on four datasets, two artificial and two real, finding that the hierarchical mixture-of-experts classifier achieves superior classification performance on the testing data.
{"title":"Using structure of data to improve classification","authors":"C. O'keefe, G. Jarrad","doi":"10.1109/IDC.2002.995419","DOIUrl":"https://doi.org/10.1109/IDC.2002.995419","url":null,"abstract":"Statistical mixture-of-experts models are often used for data analysis tasks such as clustering, regression and classification. We consider two mixture-of-experts models, the shared mixture classifier and the hierarchical mixture-of-experts classifier. We discuss the initialisation and optimisation of the structure and parameters of each classifier. In particular, we initialise the hierarchical mixture of experts classifier with the public domain OC1 decision tree software. We compare the performance of the two classifiers on four datasets, two artificial and two real, finding that the hierarchical mixture-of-experts classifier achieves superior classification performance on the testing data.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124442629","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}
Tracking of a ballistic reentry object from radar observations is a highly complex problem in nonlinear filtering. We derive the Cramer-Rao lower bounds (CRLBs) for the variance of the estimation error for this problem. Subsequently we compare several nonlinear filtering techniques to the derived CRLBs. The considered nonlinear filters include the extended Kalman filter, the unscented Kalman filter and the bootstrap (particle) filter. Considering the computational and statistical performance, the unscented Kalman filter is found to be the preferred choice for this application.
{"title":"Tracking a ballistic object on reentry: performance bounds and comparison of nonlinear filters","authors":"B. Ristic, A. Farina, D. Benvenuti","doi":"10.1109/IDC.2002.995408","DOIUrl":"https://doi.org/10.1109/IDC.2002.995408","url":null,"abstract":"Tracking of a ballistic reentry object from radar observations is a highly complex problem in nonlinear filtering. We derive the Cramer-Rao lower bounds (CRLBs) for the variance of the estimation error for this problem. Subsequently we compare several nonlinear filtering techniques to the derived CRLBs. The considered nonlinear filters include the extended Kalman filter, the unscented Kalman filter and the bootstrap (particle) filter. Considering the computational and statistical performance, the unscented Kalman filter is found to be the preferred choice for this application.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133090876","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}
In the management of large enterprise data communication networks, it is difficult to discern the causes of abnormal variations in traffic distributions on the physical network because the underlying logical topology can be very dynamic. This paper uses several graph distance measures to explore whether abnormal changes in logical connectivity and traffic flows can be detected in a sample of real network traffic. Traffic parameters were collected from a network probe attached to a physical link within a large enterprise data network. Localisation techniques were also examined to show where in the logical network most changes occurred.
{"title":"Measuring change in large enterprise data networks","authors":"H. Bunke, M. Kraetzl, P. Shoubridge, W. Wallis","doi":"10.1109/IDC.2002.995365","DOIUrl":"https://doi.org/10.1109/IDC.2002.995365","url":null,"abstract":"In the management of large enterprise data communication networks, it is difficult to discern the causes of abnormal variations in traffic distributions on the physical network because the underlying logical topology can be very dynamic. This paper uses several graph distance measures to explore whether abnormal changes in logical connectivity and traffic flows can be detected in a sample of real network traffic. Traffic parameters were collected from a network probe attached to a physical link within a large enterprise data network. Localisation techniques were also examined to show where in the logical network most changes occurred.","PeriodicalId":385351,"journal":{"name":"Final Program and Abstracts on Information, Decision and Control","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123464810","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：