Pub Date : 1900-01-01DOI: 10.1109/ISIC.1999.796667
J. Piater, R. Grupen, K. Ramamritham
Online learning robotic systems have many desirable properties. This work contributes a reinforcement learning framework for learning a time-constrained closed-loop control policy. The task is to verge the two cameras of a stereo vision system to foveate on the same world feature, within a limited number of perception-action cycles. Online learning is beneficial in at least the following ways: 1) the control parameters are optimized with respect to the characteristics of the environment actually encountered during operation; 2) visual feedback contributes to the choice of the best control action at every step in a multi-step control policy; 3) no initial calibration or explicit modeling of system parameters is required; and 4) the system can be made to adapt to non-stationary environments. Our vergence system provides a running estimate of the resulting verge quality that can be exploited by a real-time scheduler. It is shown to perform superior to two hand-calibrated vergence policies.
{"title":"Learning real-time stereo vergence control","authors":"J. Piater, R. Grupen, K. Ramamritham","doi":"10.1109/ISIC.1999.796667","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796667","url":null,"abstract":"Online learning robotic systems have many desirable properties. This work contributes a reinforcement learning framework for learning a time-constrained closed-loop control policy. The task is to verge the two cameras of a stereo vision system to foveate on the same world feature, within a limited number of perception-action cycles. Online learning is beneficial in at least the following ways: 1) the control parameters are optimized with respect to the characteristics of the environment actually encountered during operation; 2) visual feedback contributes to the choice of the best control action at every step in a multi-step control policy; 3) no initial calibration or explicit modeling of system parameters is required; and 4) the system can be made to adapt to non-stationary environments. Our vergence system provides a running estimate of the resulting verge quality that can be exploited by a real-time scheduler. It is shown to perform superior to two hand-calibrated vergence policies.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"105 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116609962","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796672
K. Moore, N. Flann
Describes a multi-resolution behavior generation strategy for a novel six-wheel omni-directional autonomous robot. The strategy is characterized by a hierarchical task decomposition approach. At the supervisory level a knowledge-based planner and an A*-optimization algorithm are used to specify the vehicle's path as a sequence of basic maneuvers. At the vehicle level these basic maneuvers are converted to time-domain trajectories. These trajectories are then tracked in an inertial reference frame using a model-based feedback linearization controller that computes set points for each wheel's low-level drive motor and steering angle motor controllers. The effectiveness of the strategy is demonstrated in actual tests with a real robot in which the path planning and control algorithms are implemented in a distributed processing environment.
{"title":"Hierarchical task decomposition approach to path planning and control for an omni-directional autonomous mobile robot","authors":"K. Moore, N. Flann","doi":"10.1109/ISIC.1999.796672","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796672","url":null,"abstract":"Describes a multi-resolution behavior generation strategy for a novel six-wheel omni-directional autonomous robot. The strategy is characterized by a hierarchical task decomposition approach. At the supervisory level a knowledge-based planner and an A*-optimization algorithm are used to specify the vehicle's path as a sequence of basic maneuvers. At the vehicle level these basic maneuvers are converted to time-domain trajectories. These trajectories are then tracked in an inertial reference frame using a model-based feedback linearization controller that computes set points for each wheel's low-level drive motor and steering angle motor controllers. The effectiveness of the strategy is demonstrated in actual tests with a real robot in which the path planning and control algorithms are implemented in a distributed processing environment.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117107197","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796693
Z.X. Han, B. Walcott, G. Feng, Y.M. Zhang
This paper presents dynamic output feedback H/sup /spl infin// controller design for fuzzy dynamic systems. Three kinds of controller design methods are proposed based on a smooth Lyapunov function or a piecewise smooth Lyapunov function. The control laws are obtained by using LMI techniques and the global stability with disturbance attenuation of the closed-loop fuzzy control system is established.
{"title":"Dynamic output feedback H/sup /spl infin// controller design for fuzzy dynamic systems","authors":"Z.X. Han, B. Walcott, G. Feng, Y.M. Zhang","doi":"10.1109/ISIC.1999.796693","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796693","url":null,"abstract":"This paper presents dynamic output feedback H/sup /spl infin// controller design for fuzzy dynamic systems. Three kinds of controller design methods are proposed based on a smooth Lyapunov function or a piecewise smooth Lyapunov function. The control laws are obtained by using LMI techniques and the global stability with disturbance attenuation of the closed-loop fuzzy control system is established.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115596821","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796650
K. Li, E. Kosmatopoulos, Petros A. Ioannou, H. Boussalis
We consider the centralized, decentralized and overlapping approaches to designing controllers for shape control of a segmented telescope. In each case, we design controllers using the H/sub /spl infin// robust control approach. In the centralized control case, the disturbances are rejected over a wide frequency range and all performance requirements are met. The order of the centralized controller, however, is high, which makes it difficult to implement in real time. The decentralized control approach greatly reduces the computational and hardware requirements at the expense of performance deterioration. The overlapping approach is shown to be a trade-off between the centralized and decentralized one and provides sufficient flexibility to meet both performance and computational requirements.
{"title":"Centralized, decentralized, and overlapping control designs for a segmented telescope","authors":"K. Li, E. Kosmatopoulos, Petros A. Ioannou, H. Boussalis","doi":"10.1109/ISIC.1999.796650","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796650","url":null,"abstract":"We consider the centralized, decentralized and overlapping approaches to designing controllers for shape control of a segmented telescope. In each case, we design controllers using the H/sub /spl infin// robust control approach. In the centralized control case, the disturbances are rejected over a wide frequency range and all performance requirements are met. The order of the centralized controller, however, is high, which makes it difficult to implement in real time. The decentralized control approach greatly reduces the computational and hardware requirements at the expense of performance deterioration. The overlapping approach is shown to be a trade-off between the centralized and decentralized one and provides sufficient flexibility to meet both performance and computational requirements.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128346354","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796670
S. Aknine
Efficiency is a major concern for all problem solving systems. One way of achieving efficiency in multi-agent systems is the application of cooperation techniques to speed up problem solving. Coalition formation can improve the performances of a multi-agent system by solving problems collectively. Given the different work done on coalition formation, unfortunately, many approaches present several gaps in the definition of the agent behavioral algorithms for coalition formation. The research presented in the article outlines the problems found in these approaches and offers a solution to this problem by using preference models.
{"title":"A reliable algorithm for multi-agent coalition formation","authors":"S. Aknine","doi":"10.1109/ISIC.1999.796670","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796670","url":null,"abstract":"Efficiency is a major concern for all problem solving systems. One way of achieving efficiency in multi-agent systems is the application of cooperation techniques to speed up problem solving. Coalition formation can improve the performances of a multi-agent system by solving problems collectively. Given the different work done on coalition formation, unfortunately, many approaches present several gaps in the definition of the agent behavioral algorithms for coalition formation. The research presented in the article outlines the problems found in these approaches and offers a solution to this problem by using preference models.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127803673","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796682
C. Hagerty, I. Muchnik, C. Kulikowski, Sung-Hou Kim
We have developed a novel, general technique involving nonlinear transformation that summarizes the four hundred and two amino acid properties contained in the Nakai database in the form of two simply visualizable indices. The technique yields a higher fidelity of representation of the original data than linear methods, and results in a distribution of the amino acids which illustrates their dependence on size and hydrophobicity properties on the one hand, and their preference for belonging to loops vs. strands, on the other.
{"title":"Two indices can approximate four hundred and two amino acid properties","authors":"C. Hagerty, I. Muchnik, C. Kulikowski, Sung-Hou Kim","doi":"10.1109/ISIC.1999.796682","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796682","url":null,"abstract":"We have developed a novel, general technique involving nonlinear transformation that summarizes the four hundred and two amino acid properties contained in the Nakai database in the form of two simply visualizable indices. The technique yields a higher fidelity of representation of the original data than linear methods, and results in a distribution of the amino acids which illustrates their dependence on size and hydrophobicity properties on the one hand, and their preference for belonging to loops vs. strands, on the other.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126815477","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796683
A. Meystel
The phenomenon of autonomy is equivalent to assigning and producing a system of complicated goals that frequently form a multiresolutional hierarchy stemming from the concept of survival. Thus, autonomy of a system dwells upon the multiresolutional (multigranular, multiscale) world representation within the intelligent system and upon the associated computational mechanisms that enable the existence of MR-representation and its use. The issue of realization and/or rationality of autonomous intelligent control systems is a complicated one. It seems to be typical for animated nature but as far as engineering of autonomous systems are concerned, not all premises of the concept of autonomy are universally accepted by the scientific community. In the paper, the topics are discussed, which require additional research before the persuasive theory can be introduced.
{"title":"Theoretical aspects of multiresolutional autonomy","authors":"A. Meystel","doi":"10.1109/ISIC.1999.796683","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796683","url":null,"abstract":"The phenomenon of autonomy is equivalent to assigning and producing a system of complicated goals that frequently form a multiresolutional hierarchy stemming from the concept of survival. Thus, autonomy of a system dwells upon the multiresolutional (multigranular, multiscale) world representation within the intelligent system and upon the associated computational mechanisms that enable the existence of MR-representation and its use. The issue of realization and/or rationality of autonomous intelligent control systems is a complicated one. It seems to be typical for animated nature but as far as engineering of autonomous systems are concerned, not all premises of the concept of autonomy are universally accepted by the scientific community. In the paper, the topics are discussed, which require additional research before the persuasive theory can be introduced.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122254542","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796671
V. Vagin, A. P. Yeremeyev
Describes designing the dynamic decision support systems (DDSSs) of a semiotic type. A DDSS of such a type is oriented towards open and dynamic problem domains and it is defined by a set of various models. The example of a DDSS prototype for monitoring and management of a nuclear power block, implemented on the basis of the tools G2+GDA, is given.
{"title":"Designing the dynamic decision support systems of a semiotic type","authors":"V. Vagin, A. P. Yeremeyev","doi":"10.1109/ISIC.1999.796671","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796671","url":null,"abstract":"Describes designing the dynamic decision support systems (DDSSs) of a semiotic type. A DDSS of such a type is oriented towards open and dynamic problem domains and it is defined by a set of various models. The example of a DDSS prototype for monitoring and management of a nuclear power block, implemented on the basis of the tools G2+GDA, is given.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"352 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125632044","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796655
B. Mirkin, I. Muchnik
A method for structural clustering is proposed based on information about entity-to-set interactions that can be easily collected in digitalized images. The method is based on the multiscale structure of the problem of maximization of a so-called bottleneck link function, which can be globally solved with a greedy-wise algorithm. Extension of the construction to clusters consisting of "hard" cores and dispersed "shells" is proposed.
{"title":"Link and bottleneck link set functions and multiresolution structural clusters","authors":"B. Mirkin, I. Muchnik","doi":"10.1109/ISIC.1999.796655","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796655","url":null,"abstract":"A method for structural clustering is proposed based on information about entity-to-set interactions that can be easily collected in digitalized images. The method is based on the multiscale structure of the problem of maximization of a so-called bottleneck link function, which can be globally solved with a greedy-wise algorithm. Extension of the construction to clusters consisting of \"hard\" cores and dispersed \"shells\" is proposed.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125649871","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 : 1900-01-01DOI: 10.1109/ISIC.1999.796663
P. Tsui, O. Basir
An active vision approach is proposed for 3D object motion estimation. The motion estimation problem is formulated as the problem of planning the poses of a moving vision system so as to minimize the estimation uncertainties. A Kalman filter is employed to estimate the object motion parameters. The Riccati equation of the filter is developed as a function of the vision system control parameters, namely, position, orientation, velocity, and acceleration. This allows for the estimation uncertainties to be treated as an evolutionary process which is controlled by the vision system parameters. An objective function is formulated based on the solution of the Riccati equation to map the sensor parameters into an index of uncertainty performance. A genetic algorithm is used to search for the optimum parameters which minimize the objective function. To achieve real-time motion estimation performance an artificial neural network is proposed to relax the computational demands associated with solving the Riccati equation. Experiments to demonstrate the speed and accuracy of object motion estimation achieved by a vision system using this control scheme is discussed.
{"title":"A neural network based vision system for 3D motion estimations","authors":"P. Tsui, O. Basir","doi":"10.1109/ISIC.1999.796663","DOIUrl":"https://doi.org/10.1109/ISIC.1999.796663","url":null,"abstract":"An active vision approach is proposed for 3D object motion estimation. The motion estimation problem is formulated as the problem of planning the poses of a moving vision system so as to minimize the estimation uncertainties. A Kalman filter is employed to estimate the object motion parameters. The Riccati equation of the filter is developed as a function of the vision system control parameters, namely, position, orientation, velocity, and acceleration. This allows for the estimation uncertainties to be treated as an evolutionary process which is controlled by the vision system parameters. An objective function is formulated based on the solution of the Riccati equation to map the sensor parameters into an index of uncertainty performance. A genetic algorithm is used to search for the optimum parameters which minimize the objective function. To achieve real-time motion estimation performance an artificial neural network is proposed to relax the computational demands associated with solving the Riccati equation. Experiments to demonstrate the speed and accuracy of object motion estimation achieved by a vision system using this control scheme is discussed.","PeriodicalId":300130,"journal":{"name":"Proceedings of the 1999 IEEE International Symposium on Intelligent Control Intelligent Systems and Semiotics (Cat. No.99CH37014)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133951402","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：