Pub Date : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246949
A. V. Sureshbabu, M. Maggiali, G. Metta, A. Parmiggiani
This paper outlines the design of the hand of the R1 humanoid robot. The hand uses a completely plastic structure with embedded electronics. It has 2 actuated degrees of freedom (DOF) with 4 phalanges, coupling two phalanges to each degree of actuation. A novel series elastic module was developed within the hand. It is used in force sensing and protects the hand from impact loads. The series elastic module is designed, characterized and evaluated across the working range of the hand. The hand also has position sensors at all joints and tactile sensors for tactile feedback on its phalanges. The hand is completely self-contained with all control boards and motors housed within the structure. It is then tested and evaluated against user needs.
{"title":"Design of a force sensing hand for the R1 humanoid robot","authors":"A. V. Sureshbabu, M. Maggiali, G. Metta, A. Parmiggiani","doi":"10.1109/HUMANOIDS.2017.8246949","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246949","url":null,"abstract":"This paper outlines the design of the hand of the R1 humanoid robot. The hand uses a completely plastic structure with embedded electronics. It has 2 actuated degrees of freedom (DOF) with 4 phalanges, coupling two phalanges to each degree of actuation. A novel series elastic module was developed within the hand. It is used in force sensing and protects the hand from impact loads. The series elastic module is designed, characterized and evaluated across the working range of the hand. The hand also has position sensors at all joints and tactile sensors for tactile feedback on its phalanges. The hand is completely self-contained with all control boards and motors housed within the structure. It is then tested and evaluated against user needs.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129226624","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246932
Brandon J. DeHart, D. Kulić
There are two main goals for any mobile, bipedal system: locomotion and balance. These behaviors both require the biped to effectively move its center of mass (COM). In this work, we define an optimization framework which can be used to design a biped that maximizes its ability to move its COM, without having to define an associated controller or trajectory. We use angular momentum gain in our objective function, a measure of how efficiently a system can move its COM based on its physical properties. As a comparison, we also optimize the model using a cost of transport-based objective function over a set of trajectories and show that it provides similar results. However, the cost of transport calculation requires slow hybrid dynamics equations and hand-designed trajectories, whereas the angular momentum gain calculation requires only the joint space inertia matrix at each configuration of interest.
{"title":"Legged mechanism design with momentum gains","authors":"Brandon J. DeHart, D. Kulić","doi":"10.1109/HUMANOIDS.2017.8246932","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246932","url":null,"abstract":"There are two main goals for any mobile, bipedal system: locomotion and balance. These behaviors both require the biped to effectively move its center of mass (COM). In this work, we define an optimization framework which can be used to design a biped that maximizes its ability to move its COM, without having to define an associated controller or trajectory. We use angular momentum gain in our objective function, a measure of how efficiently a system can move its COM based on its physical properties. As a comparison, we also optimize the model using a cost of transport-based objective function over a set of trajectories and show that it provides similar results. However, the cost of transport calculation requires slow hybrid dynamics equations and hand-designed trajectories, whereas the angular momentum gain calculation requires only the joint space inertia matrix at each configuration of interest.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123415126","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246875
Daniel Tanneberg, Jan Peters, E. Rückert
Autonomous robots need to interact with unknown and unstructured environments. For continuous online adaptation in lifelong learning scenarios, they need sample-efficient mechanisms to adapt to changing environments, constraints, tasks and capabilities. In this paper, we introduce a framework for online motion planning and adaptation based on a bio-inspired stochastic recurrent neural network. By using the intrinsic motivation signal cognitive dissonance with a mental replay strategy, the robot can learn from few physical interactions and can therefore adapt to novel environments in seconds. We evaluate our online planning and adaptation framework on a KUKA LWR arm. The efficient online adaptation is shown by learning unknown workspace constraints sample-efficient within few seconds while following given via points.
{"title":"Efficient online adaptation with stochastic recurrent neural networks","authors":"Daniel Tanneberg, Jan Peters, E. Rückert","doi":"10.1109/HUMANOIDS.2017.8246875","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246875","url":null,"abstract":"Autonomous robots need to interact with unknown and unstructured environments. For continuous online adaptation in lifelong learning scenarios, they need sample-efficient mechanisms to adapt to changing environments, constraints, tasks and capabilities. In this paper, we introduce a framework for online motion planning and adaptation based on a bio-inspired stochastic recurrent neural network. By using the intrinsic motivation signal cognitive dissonance with a mental replay strategy, the robot can learn from few physical interactions and can therefore adapt to novel environments in seconds. We evaluate our online planning and adaptation framework on a KUKA LWR arm. The efficient online adaptation is shown by learning unknown workspace constraints sample-efficient within few seconds while following given via points.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123078474","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246934
J. Furukawa, Asuka Takai, J. Morimoto
In this study, we propose a databasedriven torque estimation approach for EMG-based robot control. For conventional EMG-based controllers, torque estimation models need to be carefully calibrated to control robots that have multiple degrees of freedom. However, such a calibration procedure requires significant effort and restricts the applications of EMG-based methods to practical situations. To cope with this issue, we use large-scale data acquired from other users to avoid the calibration process and propose collaborative filtering to estimate the joint torque of a new user by exploiting the previously derived relationships between the EMG signals and the joint torque of other users. To validate our proposed method, we compared the joint torque estimation performance with a standard linear conversion model. In our experiments, we controlled an upper-limb exoskeleton robot with the estimated joint torque where we used 16-ch electrodes to measure the EMG signals of subjects. In a comparison, our proposed method showed comparable control performance with the standard approach that requires a careful calibration process.
{"title":"Database-driven approach for Biosignal-based robot control with collaborative filtering","authors":"J. Furukawa, Asuka Takai, J. Morimoto","doi":"10.1109/HUMANOIDS.2017.8246934","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246934","url":null,"abstract":"In this study, we propose a databasedriven torque estimation approach for EMG-based robot control. For conventional EMG-based controllers, torque estimation models need to be carefully calibrated to control robots that have multiple degrees of freedom. However, such a calibration procedure requires significant effort and restricts the applications of EMG-based methods to practical situations. To cope with this issue, we use large-scale data acquired from other users to avoid the calibration process and propose collaborative filtering to estimate the joint torque of a new user by exploiting the previously derived relationships between the EMG signals and the joint torque of other users. To validate our proposed method, we compared the joint torque estimation performance with a standard linear conversion model. In our experiments, we controlled an upper-limb exoskeleton robot with the estimated joint torque where we used 16-ch electrodes to measure the EMG signals of subjects. In a comparison, our proposed method showed comparable control performance with the standard approach that requires a careful calibration process.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126063633","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246972
Robin Rasch, S. Wachsmuth, Matthias König
To enable personal robots to operate in human spaces, it is necessary that robots support everyday tasks like handing over an object. Studies show that robots have to move and behave human-like, to improve social acceptance. Therefore, it is necessary to study and model human movements. This paper studies and analyses the movements of arms during hand-over between two persons in order to extract the characteristic features (elementary movements of joints, duration, angular and linear velocities, etc.). In the present study, we are using inertial measurement units with 6-axis (gyroscope and accelerometer) on wrist, elbow and shoulder to measure the movements and evaluate them. Our results show a general movement pattern for hand-overs between humans with two variants of twisting the elbow. The results of our study provide a basis for developing a human-like handover controller for humanoid robot systems or human like manipulators.
{"title":"Understanding movements of hand-over between two persons to improve humanoid robot systems","authors":"Robin Rasch, S. Wachsmuth, Matthias König","doi":"10.1109/HUMANOIDS.2017.8246972","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246972","url":null,"abstract":"To enable personal robots to operate in human spaces, it is necessary that robots support everyday tasks like handing over an object. Studies show that robots have to move and behave human-like, to improve social acceptance. Therefore, it is necessary to study and model human movements. This paper studies and analyses the movements of arms during hand-over between two persons in order to extract the characteristic features (elementary movements of joints, duration, angular and linear velocities, etc.). In the present study, we are using inertial measurement units with 6-axis (gyroscope and accelerometer) on wrist, elbow and shoulder to measure the movements and evaluate them. Our results show a general movement pattern for hand-overs between humans with two variants of twisting the elbow. The results of our study provide a basis for developing a human-like handover controller for humanoid robot systems or human like manipulators.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126991603","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8239534
Tobia Marcucci, Robin Deits, M. Gabiccini, A. Bicchi, Russ Tedrake
Feedback control of robotic systems interacting with the environment through contacts is a central topic in legged robotics. One of the main challenges posed by this problem is the choice of a model sufficiently complex to capture the discontinuous nature of the dynamics but simple enough to allow online computations. Linear models have proved to be the most effective and reliable choice for smooth systems; we believe that piecewise affine (PWA) models represent their natural extension when contact phenomena occur. Discrete-time PWA systems have been deeply analyzed in the field of hybrid Model Predictive Control (MPC), but the straightforward application of MPC techniques to complex systems, such as a humanoid robot, leads to mixed-integer optimization problems which are not solvable at real-time rates. Explicit MPC methods can construct the entire control policy offline, but the resulting policy becomes too complex to compute for systems at the scale of a humanoid robot. In this paper we propose a novel algorithm which splits the computational burden between an offline sampling phase and a limited number of online convex optimizations, enabling the application of hybrid predictive controllers to higher-dimensional systems. In doing so we are willing to partially sacrifice feedback optimality, but we set stability of the system as an inviolable requirement. Simulation results of a simple planar humanoid that balances by making contact with its environment are presented to validate the proposed controller.
{"title":"Approximate hybrid model predictive control for multi-contact push recovery in complex environments","authors":"Tobia Marcucci, Robin Deits, M. Gabiccini, A. Bicchi, Russ Tedrake","doi":"10.1109/HUMANOIDS.2017.8239534","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239534","url":null,"abstract":"Feedback control of robotic systems interacting with the environment through contacts is a central topic in legged robotics. One of the main challenges posed by this problem is the choice of a model sufficiently complex to capture the discontinuous nature of the dynamics but simple enough to allow online computations. Linear models have proved to be the most effective and reliable choice for smooth systems; we believe that piecewise affine (PWA) models represent their natural extension when contact phenomena occur. Discrete-time PWA systems have been deeply analyzed in the field of hybrid Model Predictive Control (MPC), but the straightforward application of MPC techniques to complex systems, such as a humanoid robot, leads to mixed-integer optimization problems which are not solvable at real-time rates. Explicit MPC methods can construct the entire control policy offline, but the resulting policy becomes too complex to compute for systems at the scale of a humanoid robot. In this paper we propose a novel algorithm which splits the computational burden between an offline sampling phase and a limited number of online convex optimizations, enabling the application of hybrid predictive controllers to higher-dimensional systems. In doing so we are willing to partially sacrifice feedback optimality, but we set stability of the system as an inviolable requirement. Simulation results of a simple planar humanoid that balances by making contact with its environment are presented to validate the proposed controller.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130345051","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8239553
L. Steffan, Lukas Kaul, T. Asfour
Distinguishing between dynamically stable and unstable body poses during the execution of whole-body motions is of equal importance for humanoid robots and humans assisted by robotic exoskeletons. In this work, we present a study for developing a real-time system for detecting dynamic instability based on a small number of body-mounted inertial measurement units (IMUs). To this end, we systematically evaluate different online capable classifiers, operating on the data of 1 to 6 body mounted sensors, trained on a dataset of 50 disturbed motions with nearly 30,000 motion frames recorded at 100 Hz. In contrast to the majority of related studies, our system does not make use of thresholding certain sensor values but instead uses machine learning techniques to detect characteristics and patterns of features of unstable movements. We show that the right combination of classification method and sensor placement on the human body leads to very good detection results with only 3 sensors.
{"title":"Online stability estimation based on inertial sensor data for human and humanoid fall prevention","authors":"L. Steffan, Lukas Kaul, T. Asfour","doi":"10.1109/HUMANOIDS.2017.8239553","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239553","url":null,"abstract":"Distinguishing between dynamically stable and unstable body poses during the execution of whole-body motions is of equal importance for humanoid robots and humans assisted by robotic exoskeletons. In this work, we present a study for developing a real-time system for detecting dynamic instability based on a small number of body-mounted inertial measurement units (IMUs). To this end, we systematically evaluate different online capable classifiers, operating on the data of 1 to 6 body mounted sensors, trained on a dataset of 50 disturbed motions with nearly 30,000 motion frames recorded at 100 Hz. In contrast to the majority of related studies, our system does not make use of thresholding certain sensor values but instead uses machine learning techniques to detect characteristics and patterns of features of unstable movements. We show that the right combination of classification method and sensor placement on the human body leads to very good detection results with only 3 sensors.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134292384","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8239548
Erfan Shahriari, Aljaz Kramberger, A. Gams, A. Ude, S. Haddadin
In this paper, we develop a framework to encode demonstrated trajectories as periodic dynamic motion primitives (DMP) for an impedance-controlled robot and their modification to fulfil the task objective, i. e. to adapt based on the force feedback and encoded desired wrench profile via an admittance controller. This behavior by itself can violate stability. Therefore, a passivity analysis for the whole system is presented, and based on input power ports and the demonstrated reference power, a passivity observer (PO) is designed. Subsequently, a DMP phase altering law is introduced according to the passivity criterion in order to adjust the phase based on the passivity criterion. However, since this does not necessarily guarantee passivity, a suitable virtual energy tank is used. Experimental results on a Kuka LWR-4 robot polishing an unknown surface underline the real world applicability the suggested controller.
{"title":"Adapting to contacts: Energy tanks and task energy for passivity-based dynamic movement primitives","authors":"Erfan Shahriari, Aljaz Kramberger, A. Gams, A. Ude, S. Haddadin","doi":"10.1109/HUMANOIDS.2017.8239548","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239548","url":null,"abstract":"In this paper, we develop a framework to encode demonstrated trajectories as periodic dynamic motion primitives (DMP) for an impedance-controlled robot and their modification to fulfil the task objective, i. e. to adapt based on the force feedback and encoded desired wrench profile via an admittance controller. This behavior by itself can violate stability. Therefore, a passivity analysis for the whole system is presented, and based on input power ports and the demonstrated reference power, a passivity observer (PO) is designed. Subsequently, a DMP phase altering law is introduced according to the passivity criterion in order to adjust the phase based on the passivity criterion. However, since this does not necessarily guarantee passivity, a suitable virtual energy tank is used. Experimental results on a Kuka LWR-4 robot polishing an unknown surface underline the real world applicability the suggested controller.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134220093","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246881
Akihiko Yamaguchi, C. Atkeson
We explore manipulation strategies that use vision-based tactile sensing. FingerVision is a vision-based tactile sensor that provides rich tactile sensation as well as proximity sensing. Although many other tactile sensing methods are expensive in terms of cost and/or processing, FingerVision is a simple and inexpensive approach. We use a transparent skin for fingers. Tracking markers placed on the skin provides contact force and torque estimates, and processing images obtained by seeing through the transparent skin provides static (pose, shape) and dynamic (slip, deformation) information. FingerVision can sense nearby objects even when there is no contact since it is vision-based. Also the slip detection is independent from contact force, which is effective even when the force is too small to measure, such as with origami objects. The results of experiments demonstrate that several manipulation strategies with FingerVision are effective. For example the robot can grasp and pick up an origami crane without crushing it. Video: https://youtu.be/L-YbxcyRghQ
{"title":"Implementing tactile behaviors using FingerVision","authors":"Akihiko Yamaguchi, C. Atkeson","doi":"10.1109/HUMANOIDS.2017.8246881","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246881","url":null,"abstract":"We explore manipulation strategies that use vision-based tactile sensing. FingerVision is a vision-based tactile sensor that provides rich tactile sensation as well as proximity sensing. Although many other tactile sensing methods are expensive in terms of cost and/or processing, FingerVision is a simple and inexpensive approach. We use a transparent skin for fingers. Tracking markers placed on the skin provides contact force and torque estimates, and processing images obtained by seeing through the transparent skin provides static (pose, shape) and dynamic (slip, deformation) information. FingerVision can sense nearby objects even when there is no contact since it is vision-based. Also the slip detection is independent from contact force, which is effective even when the force is too small to measure, such as with origami objects. The results of experiments demonstrate that several manipulation strategies with FingerVision are effective. For example the robot can grasp and pick up an origami crane without crushing it. Video: https://youtu.be/L-YbxcyRghQ","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"109 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131822264","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 : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246895
Elco Heijmink, A. Radulescu, Brahayam Pontón, Victor Barasuol, D. Caldwell, C. Semini
The successful execution of complex modern robotic tasks often relies on the correct tuning of a large number of parameters. In this paper we present a methodology for improving the performance of a trotting gait by learning the gait parameters, impedance profile and the gains of the control architecture. We show results on a set of terrains, for various speeds using a realistic simulation of a hydraulically actuated system. Our method achieves a reduction in the gait's mechanical energy consumption during locomotion of up to 26%. The simulation results are validated in experimental trials on the hardware system.
{"title":"Learning optimal gait parameters and impedance profiles for legged locomotion","authors":"Elco Heijmink, A. Radulescu, Brahayam Pontón, Victor Barasuol, D. Caldwell, C. Semini","doi":"10.1109/HUMANOIDS.2017.8246895","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246895","url":null,"abstract":"The successful execution of complex modern robotic tasks often relies on the correct tuning of a large number of parameters. In this paper we present a methodology for improving the performance of a trotting gait by learning the gait parameters, impedance profile and the gains of the control architecture. We show results on a set of terrains, for various speeds using a realistic simulation of a hydraulically actuated system. Our method achieves a reduction in the gait's mechanical energy consumption during locomotion of up to 26%. The simulation results are validated in experimental trials on the hardware system.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132911594","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：