Pub Date : 2002-08-07DOI: 10.1109/IJCNN.2002.1005522
N. Bergfeldt, F. Linåker
We show how a simple layered system can self-organize into a set of distinct states and qualitatively different behaviors as a result of the learning a robotic delayed response task. Our approach is based on an architecture where higher levels are able to dynamically modulate the lower reactive mapping when needed.
{"title":"Self-organized modulation of a neural robot controller","authors":"N. Bergfeldt, F. Linåker","doi":"10.1109/IJCNN.2002.1005522","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1005522","url":null,"abstract":"We show how a simple layered system can self-organize into a set of distinct states and qualitatively different behaviors as a result of the learning a robotic delayed response task. Our approach is based on an architecture where higher levels are able to dynamically modulate the lower reactive mapping when needed.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"69 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":"133039339","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007675
N.J. Saggioro, J. A. Cagnon, I. D. da Silva
In most of the cases, the systems of water distribution from groundwater wells use electrical submersible pumps. All electrical energy is applied to the pumps; however, other components (pipes, valves, etc.) of these systems are also responsible by the higher or lower consumption of electric energy. The supervisors and operators of the systems should thus have knowledge of the global energetic behavior of the process in order to administrate it properly. This work suggests a 'global energy efficiency indicator' for groundwater wells by using mathematical equations and neural networks. Simulation results are presented in order to demonstrate the validity of the proposed approach.
{"title":"A neural approach for determination of global energetic efficiency indicator in groundwater wells","authors":"N.J. Saggioro, J. A. Cagnon, I. D. da Silva","doi":"10.1109/IJCNN.2002.1007675","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007675","url":null,"abstract":"In most of the cases, the systems of water distribution from groundwater wells use electrical submersible pumps. All electrical energy is applied to the pumps; however, other components (pipes, valves, etc.) of these systems are also responsible by the higher or lower consumption of electric energy. The supervisors and operators of the systems should thus have knowledge of the global energetic behavior of the process in order to administrate it properly. This work suggests a 'global energy efficiency indicator' for groundwater wells by using mathematical equations and neural networks. Simulation results are presented in order to demonstrate the validity of the proposed approach.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"147 15","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133051285","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007551
Y. Osana
We propose a chaotic associative memory (CAM) using distributed patterns for image retrieval. This model is based on the CAM which can separate superimposed patterns and the multi winners self-organizing neural network which has the ability to generate distributed representation patterns corresponding to input in a self-organizing manner.
{"title":"Chaotic associative memory using distributed patterns for image retrieval","authors":"Y. Osana","doi":"10.1109/IJCNN.2002.1007551","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007551","url":null,"abstract":"We propose a chaotic associative memory (CAM) using distributed patterns for image retrieval. This model is based on the CAM which can separate superimposed patterns and the multi winners self-organizing neural network which has the ability to generate distributed representation patterns corresponding to input in a self-organizing manner.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"113 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":"133853545","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007594
A. Hanamitsu, M. Ohta
The maximum neural network (MNN) with self-feedbacks for the channel assignment problem (CAP) is proposed. The CAP is one of the extremely important problems in cellular mobile systems. The CAP is to assign a channel to each call in order to minimize the interference and use available channels efficiently. Funabiki et al. (2000) have proposed the hysteresis binary neuron model for the CAP and it can find lower bound solutions for well-known benchmark problems. In order to avoid converging to a local minimum, this model introduces the hill-climbing term and the omega function. Although these methodologies are effective to escape from a local minimum, they need to adjust many parameters. In this paper, the MNN with self-feedbacks is proposed in order to reduce parameters. Our proposal is applied to the CAP, and it is compared with the hysteresis binary neuron model. Our model can find the lower bound solutions in all of the benchmark problems and the average iteration step decreases by 55.5[%].
{"title":"A maximum neural network with self-feedbacks for channel assignment in cellular mobile systems","authors":"A. Hanamitsu, M. Ohta","doi":"10.1109/IJCNN.2002.1007594","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007594","url":null,"abstract":"The maximum neural network (MNN) with self-feedbacks for the channel assignment problem (CAP) is proposed. The CAP is one of the extremely important problems in cellular mobile systems. The CAP is to assign a channel to each call in order to minimize the interference and use available channels efficiently. Funabiki et al. (2000) have proposed the hysteresis binary neuron model for the CAP and it can find lower bound solutions for well-known benchmark problems. In order to avoid converging to a local minimum, this model introduces the hill-climbing term and the omega function. Although these methodologies are effective to escape from a local minimum, they need to adjust many parameters. In this paper, the MNN with self-feedbacks is proposed in order to reduce parameters. Our proposal is applied to the CAP, and it is compared with the hysteresis binary neuron model. Our model can find the lower bound solutions in all of the benchmark problems and the average iteration step decreases by 55.5[%].","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"40 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":"115391184","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007547
C. Chow, Tong Lee
An approach is introduced to determine the topology of a feedforward binary neural network automatically. The approach is based on a construction algorithm that constructs one layer of hidden nodes at a time until the problem is solved. In each layer, the algorithm determines the necessary number of nodes through a growth process by finding the best hidden node that would help to partition the input training data set. This is done using a genetic algorithm. The proposed algorithm can determine the necessary number of hidden layers and number of hidden nodes at each layer automatically. Tests on a number of benchmark problems illustrated the effectiveness of the proposed technique, both in terms of network complexity and recognition accuracy, compared with a geometrical learning approach.
{"title":"Construction of multi-layer feedforward binary neural network by a genetic algorithm","authors":"C. Chow, Tong Lee","doi":"10.1109/IJCNN.2002.1007547","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007547","url":null,"abstract":"An approach is introduced to determine the topology of a feedforward binary neural network automatically. The approach is based on a construction algorithm that constructs one layer of hidden nodes at a time until the problem is solved. In each layer, the algorithm determines the necessary number of nodes through a growth process by finding the best hidden node that would help to partition the input training data set. This is done using a genetic algorithm. The proposed algorithm can determine the necessary number of hidden layers and number of hidden nodes at each layer automatically. Tests on a number of benchmark problems illustrated the effectiveness of the proposed technique, both in terms of network complexity and recognition accuracy, compared with a geometrical learning approach.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"83 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":"115761087","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1005465
W. Tang, Jun Wang
A Lagrangian network which is developed from the Lagrange multiplier method, is proposed for multifingered hand grasping force optimization. The Lagrangian network is a recurrent neural network and is shown to be capable of taking into account the nonlinearity of the friction constraints between contacts. By giving the external load and the finger joint torque limits to the neural network, it asymptotically converges to a set of optimal grasping forces. Simulation results show that the proposed approach gives a better quality of optimal grasping force compared to other approaches in the literature.
{"title":"A Lagrangian network for multifingered hand grasping force optimization","authors":"W. Tang, Jun Wang","doi":"10.1109/IJCNN.2002.1005465","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1005465","url":null,"abstract":"A Lagrangian network which is developed from the Lagrange multiplier method, is proposed for multifingered hand grasping force optimization. The Lagrangian network is a recurrent neural network and is shown to be capable of taking into account the nonlinearity of the friction constraints between contacts. By giving the external load and the finger joint torque limits to the neural network, it asymptotically converges to a set of optimal grasping forces. Simulation results show that the proposed approach gives a better quality of optimal grasping force compared to other approaches in the literature.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"30 1 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":"115928781","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007662
Wei-Song Lin, C. Hung
A CMAC-based controller with a compensating neural network and an update rule is proposed to design the integral variable structure control (IVSC) of a nonlinear system. The control scheme comprises a soft supervisor controller and a CMAC neural network. Based on the Lyapunov theorem, the soft supervisor controller guarantees the global stability of the system. The CMAC neural network provides a compensatory signal to perform the equivalent control by a real-time learning algorithm. The new IVSC control scheme reduced the dependency on system parameters and eliminated the chattering of the control signal through learning. It is proved that the CMAC-based IVSC (CIVSC) scheme is globally stable in the sense that all signals involved are bounded and the tracking error will converge to zero. Simulation results of numerical example demonstrate the effectiveness and robustness of the proposed controller.
{"title":"CMAC based integral variable structure control of nonlinear system","authors":"Wei-Song Lin, C. Hung","doi":"10.1109/IJCNN.2002.1007662","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007662","url":null,"abstract":"A CMAC-based controller with a compensating neural network and an update rule is proposed to design the integral variable structure control (IVSC) of a nonlinear system. The control scheme comprises a soft supervisor controller and a CMAC neural network. Based on the Lyapunov theorem, the soft supervisor controller guarantees the global stability of the system. The CMAC neural network provides a compensatory signal to perform the equivalent control by a real-time learning algorithm. The new IVSC control scheme reduced the dependency on system parameters and eliminated the chattering of the control signal through learning. It is proved that the CMAC-based IVSC (CIVSC) scheme is globally stable in the sense that all signals involved are bounded and the tracking error will converge to zero. Simulation results of numerical example demonstrate the effectiveness and robustness of the proposed controller.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","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":"124277265","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1005563
R. Lang, K. Warwick
A novel extension to Kohonen's self-organising map, called the plastic self organising map (PSOM), is presented. PSOM is unlike any other network because it only has one phase of operation. The PSOM does not go through a training cycle before testing, like the SOM does and its variants. Each pattern is thus treated identically for all time. The algorithm uses a graph structure to represent data and can add or remove neurons to learn dynamic nonstationary pattern sets. The network is tested on a real world radar application and an artificial nonstationary problem.
{"title":"The plastic self organising map","authors":"R. Lang, K. Warwick","doi":"10.1109/IJCNN.2002.1005563","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1005563","url":null,"abstract":"A novel extension to Kohonen's self-organising map, called the plastic self organising map (PSOM), is presented. PSOM is unlike any other network because it only has one phase of operation. The PSOM does not go through a training cycle before testing, like the SOM does and its variants. Each pattern is thus treated identically for all time. The algorithm uses a graph structure to represent data and can add or remove neurons to learn dynamic nonstationary pattern sets. The network is tested on a real world radar application and an artificial nonstationary problem.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"23 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":"114915579","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007465
P. Gil, J. Henriques, P. Carvalho, H. Duarte-Ramos, A. Dourado
This paper describes the application of a nonlinear adaptive constrained model-based predictive control scheme to the distributed collector field of a solar power plant at the Plataforma Solar de Almeria (Spain). This methodology exploits the intrinsic nonlinear modelling capabilities of nonlinear state-space neural networks and their online training by means of an unscented Kalman filter. Tests on the ACUREX field illustrate the great engineering potential of the proposed control strategy.
{"title":"Adaptive neural model-based predictive control of a solar power plant","authors":"P. Gil, J. Henriques, P. Carvalho, H. Duarte-Ramos, A. Dourado","doi":"10.1109/IJCNN.2002.1007465","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007465","url":null,"abstract":"This paper describes the application of a nonlinear adaptive constrained model-based predictive control scheme to the distributed collector field of a solar power plant at the Plataforma Solar de Almeria (Spain). This methodology exploits the intrinsic nonlinear modelling capabilities of nonlinear state-space neural networks and their online training by means of an unscented Kalman filter. Tests on the ACUREX field illustrate the great engineering potential of the proposed control strategy.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","volume":"2 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":"114988740","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 : 2002-08-07DOI: 10.1109/IJCNN.2002.1007772
R. Kurozumi, S. Fujisawa, T. Yamamoto, Y. Suita
The existing method for establishing travel routes provides modeled environmental information, but it is difficult to create an environment model for the environments where electric wheelchairs travel because the environment changes constantly due to the existence of moving objects including pedestrians. In this study, we propose an automatic travelling system for an electric wheelchair using reinforcement learning systems and CMACs. We select the best travel route by utilizing these reinforcement learning systems. When a CMAC learns the value function of Q-learning, an improved learning speed is achieved by utilizing the generalizing action. CMACs enable one to reduce the time needed to select the best travel route. Using simulation, a path planning experiment was performed.
{"title":"Development of an automatic travel system for electric wheelchairs using reinforcement learning systems and CMACs","authors":"R. Kurozumi, S. Fujisawa, T. Yamamoto, Y. Suita","doi":"10.1109/IJCNN.2002.1007772","DOIUrl":"https://doi.org/10.1109/IJCNN.2002.1007772","url":null,"abstract":"The existing method for establishing travel routes provides modeled environmental information, but it is difficult to create an environment model for the environments where electric wheelchairs travel because the environment changes constantly due to the existence of moving objects including pedestrians. In this study, we propose an automatic travelling system for an electric wheelchair using reinforcement learning systems and CMACs. We select the best travel route by utilizing these reinforcement learning systems. When a CMAC learns the value function of Q-learning, an improved learning speed is achieved by utilizing the generalizing action. CMACs enable one to reduce the time needed to select the best travel route. Using simulation, a path planning experiment was performed.","PeriodicalId":382771,"journal":{"name":"Proceedings of the 2002 International Joint Conference on Neural Networks. IJCNN'02 (Cat. No.02CH37290)","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":"115029381","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：