Pub Date : 2008-05-19DOI: 10.1109/ROBOT.2008.4543493
Wenjie Dong, J. Farrell
This paper considers the design of control laws for multiple mechanical systems with parameter uncertainty such that the state of each system converges to a point which moves along a desired trajectory. Adaptive cooperative control laws are proposed with the aid of the passivity property of system dynamics and the results for graph theory. As an application, the proposed results are used to solve the cooperative control problem of multiple mobile robots with parameter uncertainty such that multiple mobile robots converge to a desired pattern which moves along a desired trajectory. To show effectiveness of the proposed results, simulation results are presented.
{"title":"Consensus of multiple uncertain mechanical systems and its application in cooperative control of mobile robots","authors":"Wenjie Dong, J. Farrell","doi":"10.1109/ROBOT.2008.4543493","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543493","url":null,"abstract":"This paper considers the design of control laws for multiple mechanical systems with parameter uncertainty such that the state of each system converges to a point which moves along a desired trajectory. Adaptive cooperative control laws are proposed with the aid of the passivity property of system dynamics and the results for graph theory. As an application, the proposed results are used to solve the cooperative control problem of multiple mobile robots with parameter uncertainty such that multiple mobile robots converge to a desired pattern which moves along a desired trajectory. To show effectiveness of the proposed results, simulation results are presented.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115815314","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543683
H. Uvet, T. Arai, Y. Mae, T. Takubo
This paper describes research and development of a compact vision system for real-time cell detection and manipulation in a capillary in order to apply on the "automated nuclear transplantation" project which will achieve sort of cell manipulation tasks such as positioning, cutting, sizing, and so on, automatically between different interconnected modules. Here, we propose a vision system boarded on a PDMS based silicon chip, which can be utilized in a complex network for continuous monitoring of mammalian egg and donor cells of sizes in the range of 10 to 100 micron. The developed prototype has sufficient resolution and is accompanied with a robust detection method for cell-based microfluidic applications.
{"title":"Development of a compact vision system for “automated nuclear transplantation project”","authors":"H. Uvet, T. Arai, Y. Mae, T. Takubo","doi":"10.1109/ROBOT.2008.4543683","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543683","url":null,"abstract":"This paper describes research and development of a compact vision system for real-time cell detection and manipulation in a capillary in order to apply on the \"automated nuclear transplantation\" project which will achieve sort of cell manipulation tasks such as positioning, cutting, sizing, and so on, automatically between different interconnected modules. Here, we propose a vision system boarded on a PDMS based silicon chip, which can be utilized in a complex network for continuous monitoring of mammalian egg and donor cells of sizes in the range of 10 to 100 micron. The developed prototype has sufficient resolution and is accompanied with a robust detection method for cell-based microfluidic applications.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116909375","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543621
J. Morimoto, S. Hyon, C. Atkeson, G. Cheng
We propose using the kernel dimension reduction (KDR) to extract a low-dimensional feature space for humanoid locomotion tasks. Although humanoids have many degrees of freedom, task relevant feature spaces can be much smaller than the number of dimension of the original state space. We consider an application of the proposed approach to improve the locomotive performance of humanoid robots using an extracted low-dimensional state space. To improve the locomotive performance, we use a reinforcement learning (RL) framework. While RL is a useful non-linear optimizer, it is usually difficult to apply RL to real robotic systems - due to the large number of iterations required to acquire suitable policies. In this study, we use the extracted low-dimensional feature space for RL so that the learning system can improve task performance quickly. The kernel dimension reduction method allows us to extract the feature space even if the task relevant mapping is non-linear. This is an essential property to improve humanoid locomotive performance since stepping or walking dynamics involves highly nonlinear dynamics. We show that we can improve stepping and walking policies by using a RL method on an extracted feature space by using KDR.
{"title":"Low-dimensional feature extraction for humanoid locomotion using kernel dimension reduction","authors":"J. Morimoto, S. Hyon, C. Atkeson, G. Cheng","doi":"10.1109/ROBOT.2008.4543621","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543621","url":null,"abstract":"We propose using the kernel dimension reduction (KDR) to extract a low-dimensional feature space for humanoid locomotion tasks. Although humanoids have many degrees of freedom, task relevant feature spaces can be much smaller than the number of dimension of the original state space. We consider an application of the proposed approach to improve the locomotive performance of humanoid robots using an extracted low-dimensional state space. To improve the locomotive performance, we use a reinforcement learning (RL) framework. While RL is a useful non-linear optimizer, it is usually difficult to apply RL to real robotic systems - due to the large number of iterations required to acquire suitable policies. In this study, we use the extracted low-dimensional feature space for RL so that the learning system can improve task performance quickly. The kernel dimension reduction method allows us to extract the feature space even if the task relevant mapping is non-linear. This is an essential property to improve humanoid locomotive performance since stepping or walking dynamics involves highly nonlinear dynamics. We show that we can improve stepping and walking policies by using a RL method on an extracted feature space by using KDR.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116925795","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543590
Rebecca Kokes, K. Lister, R. Gullapalli, Bao Zhang, H. Richard, J. Desai
This paper presents an initial design and feasibility study for a 1-DOF Magnetic Resonance Imaging (MRI) compatible needle driver robot for radiofrequency ablation (RFA). This initial design and study is necessary to further understand how to improve on many of the shortcomings in the standard RFA procedure. Combining needle driving with advanced image tracking techniques could provide improved solutions to these clinical limitations. In this paper, we present a hydraulically-actuated 1-DOF needle driver robot that is capable of advancing a radiofrequency (RF) probe into tissue at controllable velocities and positions within an MRI scanner, while collecting force feedback data and maintaining all standards of MRI-compatible design. We also present a method of interfacing the robot with a PHANToM haptic feedback device controlled from outside the MRI scanning room. Experiments demonstrating the PHANToM's ability to receive force feedback and guide the RFA tool to a tumor nodule within a phantom breast model while continuously imaging within MRI have been presented. Our haptic feedback system enabled us to detect normal vs. tumor phantom tissue in the preliminary experiments. Our experimental results demonstrate the compatibility of the entire system for operation during continuous MRI imaging.
{"title":"Towards a needle driver robot for radiofrequency ablation of tumors under continuous MRI","authors":"Rebecca Kokes, K. Lister, R. Gullapalli, Bao Zhang, H. Richard, J. Desai","doi":"10.1109/ROBOT.2008.4543590","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543590","url":null,"abstract":"This paper presents an initial design and feasibility study for a 1-DOF Magnetic Resonance Imaging (MRI) compatible needle driver robot for radiofrequency ablation (RFA). This initial design and study is necessary to further understand how to improve on many of the shortcomings in the standard RFA procedure. Combining needle driving with advanced image tracking techniques could provide improved solutions to these clinical limitations. In this paper, we present a hydraulically-actuated 1-DOF needle driver robot that is capable of advancing a radiofrequency (RF) probe into tissue at controllable velocities and positions within an MRI scanner, while collecting force feedback data and maintaining all standards of MRI-compatible design. We also present a method of interfacing the robot with a PHANToM haptic feedback device controlled from outside the MRI scanning room. Experiments demonstrating the PHANToM's ability to receive force feedback and guide the RFA tool to a tumor nodule within a phantom breast model while continuously imaging within MRI have been presented. Our haptic feedback system enabled us to detect normal vs. tumor phantom tissue in the preliminary experiments. Our experimental results demonstrate the compatibility of the entire system for operation during continuous MRI imaging.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127106545","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543219
Taro Takahashi, Toshimitsu Tsuboi, T. Kishida, Yasunori Kawanami, Satoru Shimizu, M. Iribe, T. Fukushima, M. Fujita
In this paper we propose a new robust force and position control method for property-unknown objects grasping. The proposed control method is capable of selecting the force control or position control, and smooth and quick switching according to the amount of the external force. The proposed method was applied to adaptive grasping by three-fingered hand which has 12 DOF, and the experimental results revealed that the smooth collision process and the stable grasping is realized even if the precise surface position, the mass and the stiffness are unknown. In addition a new algorithm determines the grasp force according to the "slip" measured with the tactile sensor and the viscoelastic media on the fingertip. This algorithm works at starting and stationary state, so the friction and mass unknown object grasping is realized by the effectual force.
{"title":"Adaptive grasping by multi fingered hand with tactile sensor based on robust force and position control","authors":"Taro Takahashi, Toshimitsu Tsuboi, T. Kishida, Yasunori Kawanami, Satoru Shimizu, M. Iribe, T. Fukushima, M. Fujita","doi":"10.1109/ROBOT.2008.4543219","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543219","url":null,"abstract":"In this paper we propose a new robust force and position control method for property-unknown objects grasping. The proposed control method is capable of selecting the force control or position control, and smooth and quick switching according to the amount of the external force. The proposed method was applied to adaptive grasping by three-fingered hand which has 12 DOF, and the experimental results revealed that the smooth collision process and the stable grasping is realized even if the precise surface position, the mass and the stiffness are unknown. In addition a new algorithm determines the grasp force according to the \"slip\" measured with the tactile sensor and the viscoelastic media on the fingertip. This algorithm works at starting and stationary state, so the friction and mass unknown object grasping is realized by the effectual force.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127486022","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543220
Dov Katz, O. Brock
Robust robotic manipulation and perception remains a difficult challenge, in particular in unstructured environments. To address this challenge, we propose to couple manipulation and perception. The robot observes its own deliberate interactions with the world. These interactions reveal sensory information that would otherwise remain hidden and facilitate the interpretation of perceptual data. To demonstrate the effectiveness of interactive perception we present a skill for the manipulation of articulated objects. We show how UMan, our mobile manipulation platform, obtains a kinematic model of an unknown object. The model then enables the robot to perform purposeful manipulation. Our algorithm is extremely robust, and does not require prior knowledge of the object; it is insensitive to lighting, texture, color, specularities, background, and is computationally highly efficient.
{"title":"Manipulating articulated objects with interactive perception","authors":"Dov Katz, O. Brock","doi":"10.1109/ROBOT.2008.4543220","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543220","url":null,"abstract":"Robust robotic manipulation and perception remains a difficult challenge, in particular in unstructured environments. To address this challenge, we propose to couple manipulation and perception. The robot observes its own deliberate interactions with the world. These interactions reveal sensory information that would otherwise remain hidden and facilitate the interpretation of perceptual data. To demonstrate the effectiveness of interactive perception we present a skill for the manipulation of articulated objects. We show how UMan, our mobile manipulation platform, obtains a kinematic model of an unknown object. The model then enables the robot to perform purposeful manipulation. Our algorithm is extremely robust, and does not require prior knowledge of the object; it is insensitive to lighting, texture, color, specularities, background, and is computationally highly efficient.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127505787","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543526
Matt Johnson, P. Feltovich, J. Bradshaw, L. Bunch
Several key aspects of coordination such as teamwork, roles, and communication, are enabled and driven by, and even largely defined by, various systems of regulation. One key feature of all these elements in human coordination is their dynamic nature. We have developed a framework to provide a dynamic regulatory system for supporting coordination in human-robot teamwork. This framework supports the definition and functions of roles within teams, as well as the creation of subteams and the roles within them. It also serves to regulate communications in support of coordination. We have demonstrated our system with a team of two humans and five robots performing advanced coordination while trying to apprehend an intruder hiding on a cluttered pier. This work lays the foundation for human-robot coordination based on dynamic regulation.
{"title":"Human-robot coordination through dynamic regulation","authors":"Matt Johnson, P. Feltovich, J. Bradshaw, L. Bunch","doi":"10.1109/ROBOT.2008.4543526","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543526","url":null,"abstract":"Several key aspects of coordination such as teamwork, roles, and communication, are enabled and driven by, and even largely defined by, various systems of regulation. One key feature of all these elements in human coordination is their dynamic nature. We have developed a framework to provide a dynamic regulatory system for supporting coordination in human-robot teamwork. This framework supports the definition and functions of roles within teams, as well as the creation of subteams and the roles within them. It also serves to regulate communications in support of coordination. We have demonstrated our system with a team of two humans and five robots performing advanced coordination while trying to apprehend an intruder hiding on a cluttered pier. This work lays the foundation for human-robot coordination based on dynamic regulation.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125850728","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543653
G. Kósa, P. Jakab, F. Jolesz, N. Hata
Capsule endoscopy is a promising technique for diagnosing diseases in the small intestines. Here we propose a miniature swimming mechanism that uses MRI's magnetic fields for both propulsion and wireless energy delivery. Our method uses both the static and radio frequency (RF) magnetic field inherently available in MRI to generate propulsion force. The propulsion force is produced by a swimming tail containing waving beam consisting of three coils in a row. Alternating current in the coils acting on the static magnetic field of the MRI will generate waving movement to produce a propulsion force. RF magnetic field will provide power to generate the alternating currents in the coils. We developed a theoretical model to predict sinusoidal waves produced by the waving beam using the Euler-Bernoulli beam equation and multiple- input multiple-output system were solved using antenna design theory. This numerical model predicted that the maximal propulsion from a 10 mm long tail can produce a velocity of 7.9 mm/s force of 5.5 mN when placed in a 3T static magnetic field. A validation study with a single coil demonstrated that the theoretical and numerical model predicts well the proposed swimming mechanism and it is useful for the fabrication of swimming tails.
{"title":"Swimming capsule endoscope using static and RF magnetic field of MRI for propulsion","authors":"G. Kósa, P. Jakab, F. Jolesz, N. Hata","doi":"10.1109/ROBOT.2008.4543653","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543653","url":null,"abstract":"Capsule endoscopy is a promising technique for diagnosing diseases in the small intestines. Here we propose a miniature swimming mechanism that uses MRI's magnetic fields for both propulsion and wireless energy delivery. Our method uses both the static and radio frequency (RF) magnetic field inherently available in MRI to generate propulsion force. The propulsion force is produced by a swimming tail containing waving beam consisting of three coils in a row. Alternating current in the coils acting on the static magnetic field of the MRI will generate waving movement to produce a propulsion force. RF magnetic field will provide power to generate the alternating currents in the coils. We developed a theoretical model to predict sinusoidal waves produced by the waving beam using the Euler-Bernoulli beam equation and multiple- input multiple-output system were solved using antenna design theory. This numerical model predicted that the maximal propulsion from a 10 mm long tail can produce a velocity of 7.9 mm/s force of 5.5 mN when placed in a 3T static magnetic field. A validation study with a single coil demonstrated that the theoretical and numerical model predicts well the proposed swimming mechanism and it is useful for the fabrication of swimming tails.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125890237","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543184
Andrew E. Johnson, S. Goldberg, Yang Cheng, L. Matthies
Visual odometry can augment or replace wheel odometry when navigating in high slip terrain which is quite important for autonomous navigation on Mars. We present a computationally efficient and robust visual odometry algorithm developed for the Mars Science Laboratory mission. This algorithm is a significant improvement over the algorithm developed for the Mars Exploration Rover Mission because it is at least four time more computationally efficient and it tracks significantly more features. The core of the algorithm is an integrated motion estimation and stereo feature tracking loop that allows for feature recovery while guiding feature correlation search to minimize computation. Results on thousands of terrestrial and Martian stereo pairs show that the algorithm can operate with no initial motion estimate while still obtaining subpixel attitude estimation performance.
{"title":"Robust and Efficient Stereo Feature Tracking for Visual Odometry","authors":"Andrew E. Johnson, S. Goldberg, Yang Cheng, L. Matthies","doi":"10.1109/ROBOT.2008.4543184","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543184","url":null,"abstract":"Visual odometry can augment or replace wheel odometry when navigating in high slip terrain which is quite important for autonomous navigation on Mars. We present a computationally efficient and robust visual odometry algorithm developed for the Mars Science Laboratory mission. This algorithm is a significant improvement over the algorithm developed for the Mars Exploration Rover Mission because it is at least four time more computationally efficient and it tracks significantly more features. The core of the algorithm is an integrated motion estimation and stereo feature tracking loop that allows for feature recovery while guiding feature correlation search to minimize computation. Results on thousands of terrestrial and Martian stereo pairs show that the algorithm can operate with no initial motion estimate while still obtaining subpixel attitude estimation performance.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"1198 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126192703","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 : 2008-05-19DOI: 10.1109/ROBOT.2008.4543798
W. Watanabe, Takahide Sato, A. Ishiguro
This study is intended to deal with the interplay between control and mechanical systems, and to discuss the "brain-body interaction as it should be" particularly from the viewpoint of learning. To this end, we have employed a decentralized control of a two-dimensional serpentine robot consisting of several identical body segments as a practical example. The preliminary simulation results derived indicate that the convergence of decentralized learning of locomotion control can be significantly improved even with an extremely simple learning algorithm, i.e., a gradient method, by introducing biarticular muscles compared to the one only with monoarticular muscles. This strongly suggests the fact that a certain amount of computation should be off loaded from the brain into its body, which allows robots to emerge various interesting functionalities.
{"title":"An efficient decentralized learning by exploiting biarticular muscles - A case study with a 2D serpentine robot -","authors":"W. Watanabe, Takahide Sato, A. Ishiguro","doi":"10.1109/ROBOT.2008.4543798","DOIUrl":"https://doi.org/10.1109/ROBOT.2008.4543798","url":null,"abstract":"This study is intended to deal with the interplay between control and mechanical systems, and to discuss the \"brain-body interaction as it should be\" particularly from the viewpoint of learning. To this end, we have employed a decentralized control of a two-dimensional serpentine robot consisting of several identical body segments as a practical example. The preliminary simulation results derived indicate that the convergence of decentralized learning of locomotion control can be significantly improved even with an extremely simple learning algorithm, i.e., a gradient method, by introducing biarticular muscles compared to the one only with monoarticular muscles. This strongly suggests the fact that a certain amount of computation should be off loaded from the brain into its body, which allows robots to emerge various interesting functionalities.","PeriodicalId":351230,"journal":{"name":"2008 IEEE International Conference on Robotics and Automation","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123253589","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：