Pub Date : 2004-12-01DOI: 10.1109/RAMECH.2004.1438039
Shuichi Fukunaga, Yutaka Nakamura, Kazuaki Aso, S. Ishii
A snake has large potential to move in various environments by drastically changing its 'gait' pattern, in spite of its simple body. We configured a control scheme for a snakelike robot, using a central pattern generator, and developed a learning algorithm to acquire a good control rule in a changing environment. Although the snake-like robot has a large degree of freedom, a computer simulation showed that a control rule that allows the robot to move to a target direction on an inclined ground or in a changing environment is obtained by our scheme.
{"title":"Reinforcement learning for a snake-like robot controlled by a central pattern generator","authors":"Shuichi Fukunaga, Yutaka Nakamura, Kazuaki Aso, S. Ishii","doi":"10.1109/RAMECH.2004.1438039","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438039","url":null,"abstract":"A snake has large potential to move in various environments by drastically changing its 'gait' pattern, in spite of its simple body. We configured a control scheme for a snakelike robot, using a central pattern generator, and developed a learning algorithm to acquire a good control rule in a changing environment. Although the snake-like robot has a large degree of freedom, a computer simulation showed that a control rule that allows the robot to move to a target direction on an inclined ground or in a changing environment is obtained by our scheme.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"101 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116098495","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1437999
Catherine Lyons, Bill Ommert, S. Thayer
The problem of human localization presents unique challenges, particularly in areas where GPS satellite signals may be blocked for a significant period of time. A new approach to localization based on a physical model of human motion is presented. Data is taken from inertial measurement units on the limbs, and gait analysis is used to detect and measure different modes of human motion from this data. This approach produces an accurate form of dead reckoning that is less prone to drift than current inertial navigation systems and more adaptable then current pedometry systems.
{"title":"Bipedal motion estimation with the human odometer","authors":"Catherine Lyons, Bill Ommert, S. Thayer","doi":"10.1109/RAMECH.2004.1437999","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1437999","url":null,"abstract":"The problem of human localization presents unique challenges, particularly in areas where GPS satellite signals may be blocked for a significant period of time. A new approach to localization based on a physical model of human motion is presented. Data is taken from inertial measurement units on the limbs, and gait analysis is used to detect and measure different modes of human motion from this data. This approach produces an accurate form of dead reckoning that is less prone to drift than current inertial navigation systems and more adaptable then current pedometry systems.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115112749","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438030
Chungjong Lee, A. Sowmya
Current robotic localization and SLAM algorithms are restricted to simple geometric features such as lines and corners as landmarks. The richness of the information provided by visual perception has not been fully explored. This paper presents a vision based SLAM algorithm which utilizes visual information with minimal prior assumptions.
{"title":"An approach to appearance-based simultaneous localization and map building (SLAM)","authors":"Chungjong Lee, A. Sowmya","doi":"10.1109/RAMECH.2004.1438030","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438030","url":null,"abstract":"Current robotic localization and SLAM algorithms are restricted to simple geometric features such as lines and corners as landmarks. The richness of the information provided by visual perception has not been fully explored. This paper presents a vision based SLAM algorithm which utilizes visual information with minimal prior assumptions.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122129170","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438014
G. Faria, R. Romero, Edson Prestes e Silva, M. Idiart
Potential field is a reactive method that has been used for trajectory control of mobile robots. In this method the robot behaves like a particle moving under the influence of an artificial potential produced by the target and the obstacles. This method has lower computational cost than others that utilize maps as a world model. However, one problem or this method is that it can generate regions of local minima. In these regions the driving force vanishes due to specific obstacle and goal configurations and the robot gets trapped. The use of harmonic functions for the potential calculation can solve this problem. In the harmonic function method the environment is represented by a grid, in which cells with an obstacle have their potential value set to 1 whereas cells that contain the target have their potential value set to 0. A relaxation algorithm is then employed to calculate the harmonic potential among the obstacles and the target. These two techniques have been applied in robot soccer environment to develop game strategies. Results obtained show that both techniques present better results when compared to the standard strategy or the FIRA Simulator. Further, a comparison of these techniques is presented for showing the advantages and disadvantages existing in each one of them.
{"title":"Comparing harmonic functions and potential fields in the trajectory control of mobile robots","authors":"G. Faria, R. Romero, Edson Prestes e Silva, M. Idiart","doi":"10.1109/RAMECH.2004.1438014","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438014","url":null,"abstract":"Potential field is a reactive method that has been used for trajectory control of mobile robots. In this method the robot behaves like a particle moving under the influence of an artificial potential produced by the target and the obstacles. This method has lower computational cost than others that utilize maps as a world model. However, one problem or this method is that it can generate regions of local minima. In these regions the driving force vanishes due to specific obstacle and goal configurations and the robot gets trapped. The use of harmonic functions for the potential calculation can solve this problem. In the harmonic function method the environment is represented by a grid, in which cells with an obstacle have their potential value set to 1 whereas cells that contain the target have their potential value set to 0. A relaxation algorithm is then employed to calculate the harmonic potential among the obstacles and the target. These two techniques have been applied in robot soccer environment to develop game strategies. Results obtained show that both techniques present better results when compared to the standard strategy or the FIRA Simulator. Further, a comparison of these techniques is presented for showing the advantages and disadvantages existing in each one of them.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121434085","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438976
Liang Yan, K. Chee, I. Chen, Guilin Yang, Wei Lin
A new form of 3-DOF spherical actuator having a rotor mounted with rare-earth-magnet poles is developed. Since the air-core coils are used in this actuator, the magnetic flux distribution is different from that of conventional electromagnetic devices. In this paper, the solution of the Laplace equation is used to derive the general form of the magnetic field, and a curve fitting approach is proposed to obtain the mathematical expression of the magnetic field. This mathematical representation of the 3D magnetic field can be applied to the torque modelling of the 3-DOF spherical actuator. Experimental studies have been conducted to determine the critical parameters and verify the effectiveness of the proposed approach.
{"title":"A hybrid approach for magnetic field analysis","authors":"Liang Yan, K. Chee, I. Chen, Guilin Yang, Wei Lin","doi":"10.1109/RAMECH.2004.1438976","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438976","url":null,"abstract":"A new form of 3-DOF spherical actuator having a rotor mounted with rare-earth-magnet poles is developed. Since the air-core coils are used in this actuator, the magnetic flux distribution is different from that of conventional electromagnetic devices. In this paper, the solution of the Laplace equation is used to derive the general form of the magnetic field, and a curve fitting approach is proposed to obtain the mathematical expression of the magnetic field. This mathematical representation of the 3D magnetic field can be applied to the torque modelling of the 3-DOF spherical actuator. Experimental studies have been conducted to determine the critical parameters and verify the effectiveness of the proposed approach.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"437 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122881768","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438079
Xinyuan He, Shugen Ma, Bin Li, Yuechao Wang, C. Ye
For improvement of planetary rover locomotion, a new type of tracked vehicle was designed for planetary exploration based on the concept of reconfigurability. By applying the planetary wheel principle, the system can give output in different forms under various constraint conditions. As a reconfigurable module, the vehicle can be integrated into a larger system using two basic configurations. As a stand-alone rover, it has two moving modes. The whole design was analyzed by conducting dynamics simulations and experiments.
{"title":"Analysis of locomotion of a planetary rover on a slope","authors":"Xinyuan He, Shugen Ma, Bin Li, Yuechao Wang, C. Ye","doi":"10.1109/RAMECH.2004.1438079","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438079","url":null,"abstract":"For improvement of planetary rover locomotion, a new type of tracked vehicle was designed for planetary exploration based on the concept of reconfigurability. By applying the planetary wheel principle, the system can give output in different forms under various constraint conditions. As a reconfigurable module, the vehicle can be integrated into a larger system using two basic configurations. As a stand-alone rover, it has two moving modes. The whole design was analyzed by conducting dynamics simulations and experiments.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125589338","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438013
Hiroki Miura, G. Obinata, Atsushi Nakayama, K. Hase, M. Sasaki, T. Iwami, H. Doki
We introduce an optimization method for the trajectory planning of redundant manipulators which achieve a given task with high efficiency, and apply the method to a wheelchair propulsion problem. A GA is used to optimize the redundant variables of a manipulator. Additionally, the procedure of the method does not use any forward dynamics computation. The effectiveness of the proposed algorithm has been shown by applying the method to wheelchair propulsion.
{"title":"Task based approach on trajectory planning with redundant manipulators, and its application to wheelchair propulsion","authors":"Hiroki Miura, G. Obinata, Atsushi Nakayama, K. Hase, M. Sasaki, T. Iwami, H. Doki","doi":"10.1109/RAMECH.2004.1438013","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438013","url":null,"abstract":"We introduce an optimization method for the trajectory planning of redundant manipulators which achieve a given task with high efficiency, and apply the method to a wheelchair propulsion problem. A GA is used to optimize the redundant variables of a manipulator. Additionally, the procedure of the method does not use any forward dynamics computation. The effectiveness of the proposed algorithm has been shown by applying the method to wheelchair propulsion.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"87 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129316676","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438958
P. V. D. Ven, C. Flanagan, D. Toal, E. Omerdic
In this paper the use of neural networks for the identification of underwater vehicle dynamics is studied. A flexible way of identifying dynamics is desirable for several reasons. The dynamics of underwater craft are highly non-linear and cross coupling between various degrees of freedom normally exists. To date at best empirical models are available to describe these phenomena. On top of this the underwater environment can change drastically as a result of, for example, weather conditions. Due to their ability to adapt for changing circumstances in an online fashion, neural networks offer an interesting alternative for more conventional means of identification. This paper details an identification process using neural networks. To illustrate the performance of this identification process, these neural networks are then used directly or indirectly in a feedforward loop to control the craft in a simulation study.
{"title":"Identification and control of underwater vehicles with the aid of neural networks","authors":"P. V. D. Ven, C. Flanagan, D. Toal, E. Omerdic","doi":"10.1109/RAMECH.2004.1438958","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438958","url":null,"abstract":"In this paper the use of neural networks for the identification of underwater vehicle dynamics is studied. A flexible way of identifying dynamics is desirable for several reasons. The dynamics of underwater craft are highly non-linear and cross coupling between various degrees of freedom normally exists. To date at best empirical models are available to describe these phenomena. On top of this the underwater environment can change drastically as a result of, for example, weather conditions. Due to their ability to adapt for changing circumstances in an online fashion, neural networks offer an interesting alternative for more conventional means of identification. This paper details an identification process using neural networks. To illustrate the performance of this identification process, these neural networks are then used directly or indirectly in a feedforward loop to control the craft in a simulation study.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122378901","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438047
Jie Pang, Guizhi Li, Wei Zhang, Xuyan Tu
The Internet faces the following serious problems: (1) the complexity of information, (2) serious network security loophole, (3) can't provide humanization service. For the sake of seeking a new path for the above problems, based on the works of domestic and international experts, we put forward the concept of "softman". It's basic speculation is: regarding AI (artificial intelligence), AL (artificial life) and distributed system as its theoretical foundations, integrating techniques of "intelligent robot", "intelligent network" and "MAS (multi agent system)", we can develop a kind of automatic intelligence network tool: "softman". It can migrate freely in cyberspace and can adopt the "IIPP (intelligent information push-pull)" technique to handle some appointed missions automatically. The concept, essence, theory foundations, characteristics and models of softman are discussed in this paper. Finally a realizable individual softman overall model is given.
{"title":"Study on \"virtual robot in network environment - Softman\"","authors":"Jie Pang, Guizhi Li, Wei Zhang, Xuyan Tu","doi":"10.1109/RAMECH.2004.1438047","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438047","url":null,"abstract":"The Internet faces the following serious problems: (1) the complexity of information, (2) serious network security loophole, (3) can't provide humanization service. For the sake of seeking a new path for the above problems, based on the works of domestic and international experts, we put forward the concept of \"softman\". It's basic speculation is: regarding AI (artificial intelligence), AL (artificial life) and distributed system as its theoretical foundations, integrating techniques of \"intelligent robot\", \"intelligent network\" and \"MAS (multi agent system)\", we can develop a kind of automatic intelligence network tool: \"softman\". It can migrate freely in cyberspace and can adopt the \"IIPP (intelligent information push-pull)\" technique to handle some appointed missions automatically. The concept, essence, theory foundations, characteristics and models of softman are discussed in this paper. Finally a realizable individual softman overall model is given.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116304783","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 : 2004-12-01DOI: 10.1109/RAMECH.2004.1438988
F. Lin, Po-Hung Shen
An adaptive fuzzy neural network (AFNN) control system is proposed to control the position of the mover of a field-oriented control permanent magnet linear synchronous motor (PMLSM) servo drive system to track periodic reference trajectories in this study. In the proposed AFNN control system, a FNN with accurate approximation capability is employed to approximate the unknown dynamics of the PMLSM, and a robust compensator is proposed to confront the inevitable approximation errors due to finite number of membership functions and disturbances including the friction force. The adaptive learning algorithm that can learn the parameters of the FNN on line is derived using Lyapunov stability theorem. Moreover, to relax the requirement for the value of lumped uncertainty in the robust compensator which comprises a minimum approximation error, optimal parameter vectors, higher-order terms in Taylor series and friction force, an adaptive lumped uncertainty estimation law is investigated. Furthermore, all the control algorithms are implemented in a TMS320C32 DSP-based control computer. The experimental results due to periodic reference trajectories show that the dynamic behaviors of the proposed control systems are robust with regard to uncertainties.
{"title":"A DSP-based permanent magnet linear motor servo drive using adaptive fuzzy-neural-network control","authors":"F. Lin, Po-Hung Shen","doi":"10.1109/RAMECH.2004.1438988","DOIUrl":"https://doi.org/10.1109/RAMECH.2004.1438988","url":null,"abstract":"An adaptive fuzzy neural network (AFNN) control system is proposed to control the position of the mover of a field-oriented control permanent magnet linear synchronous motor (PMLSM) servo drive system to track periodic reference trajectories in this study. In the proposed AFNN control system, a FNN with accurate approximation capability is employed to approximate the unknown dynamics of the PMLSM, and a robust compensator is proposed to confront the inevitable approximation errors due to finite number of membership functions and disturbances including the friction force. The adaptive learning algorithm that can learn the parameters of the FNN on line is derived using Lyapunov stability theorem. Moreover, to relax the requirement for the value of lumped uncertainty in the robust compensator which comprises a minimum approximation error, optimal parameter vectors, higher-order terms in Taylor series and friction force, an adaptive lumped uncertainty estimation law is investigated. Furthermore, all the control algorithms are implemented in a TMS320C32 DSP-based control computer. The experimental results due to periodic reference trajectories show that the dynamic behaviors of the proposed control systems are robust with regard to uncertainties.","PeriodicalId":252964,"journal":{"name":"IEEE Conference on Robotics, Automation and Mechatronics, 2004.","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126568677","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：