Pub Date : 2017-11-01DOI: 10.1109/HUMANOIDS.2017.8246874
Affan Pervez, Yuecheng Mao, Dongheui Lee
Dynamic Movement Primitives (DMPs) are widely used for encoding motion data. Task parameterized DMP (TP-DMP) can adapt a learned skill to different situations. Mostly a customized vision system is used to extract task specific variables. This limits the use of such systems to real world scenarios. This paper proposes a method for combining the DMP with a Convolutional Neural Network (CNN). Our approach preserves the generalization properties associated with a DMP, while the CNN learns the task specific features from the camera images. This eliminates the need to extract the task parameters, by directly utilizing the camera image during the motion reproduction. The performance of the developed approach is demonstrated through a trash cleaning task, executed with a real robot. We also show that by using the data augmentation, the learned sweeping skill can be generalized for arbitrary objects. The experiments show the robustness of our approach for several different settings.
{"title":"Learning deep movement primitives using convolutional neural networks","authors":"Affan Pervez, Yuecheng Mao, Dongheui Lee","doi":"10.1109/HUMANOIDS.2017.8246874","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246874","url":null,"abstract":"Dynamic Movement Primitives (DMPs) are widely used for encoding motion data. Task parameterized DMP (TP-DMP) can adapt a learned skill to different situations. Mostly a customized vision system is used to extract task specific variables. This limits the use of such systems to real world scenarios. This paper proposes a method for combining the DMP with a Convolutional Neural Network (CNN). Our approach preserves the generalization properties associated with a DMP, while the CNN learns the task specific features from the camera images. This eliminates the need to extract the task parameters, by directly utilizing the camera image during the motion reproduction. The performance of the developed approach is demonstrated through a trash cleaning task, executed with a real robot. We also show that by using the data augmentation, the learned sweeping skill can be generalized for arbitrary objects. The experiments show the robustness of our approach for several different settings.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127922053","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.2017.8239550
Stefan Oßwald, P. Karkowski, Maren Bennewitz
Covering a known 3D environment with a robot's camera is a commonly required task, for example in inspection and surveillance, mapping, or object search applications. In addition to the problem of finding a complete and efficient set of view points for covering the whole environment, humanoid robots also need to observe balance, energy, and kinematic constraints for reaching the desired view poses. In this paper, we approach this high-dimensional planning problem by introducing a novel inverse reachability map representation that can be used for fast pose generation and combine it with a next-best-view algorithm. We implemented our approach in ROS and tested it with a Nao robot on both simulated and real-world scenes. The experiments show that our approach enables the humanoid to efficiently cover room-sized environments with its camera.
{"title":"Efficient coverage of 3D environments with humanoid robots using inverse reachability maps","authors":"Stefan Oßwald, P. Karkowski, Maren Bennewitz","doi":"10.1109/HUMANOIDS.2017.8239550","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239550","url":null,"abstract":"Covering a known 3D environment with a robot's camera is a commonly required task, for example in inspection and surveillance, mapping, or object search applications. In addition to the problem of finding a complete and efficient set of view points for covering the whole environment, humanoid robots also need to observe balance, energy, and kinematic constraints for reaching the desired view poses. In this paper, we approach this high-dimensional planning problem by introducing a novel inverse reachability map representation that can be used for fast pose generation and combine it with a next-best-view algorithm. We implemented our approach in ROS and tested it with a Nao robot on both simulated and real-world scenes. The experiments show that our approach enables the humanoid to efficiently cover room-sized environments with its camera.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121722490","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.2017.8246890
Yiwei Wang, Xin Ye, Yezhou Yang, Wenlong Zhang
We present a framework from vision based hand movement prediction in a real-world human-robot collaborative scenario for safety guarantee. We first propose a perception submodule that takes in visual data solely and predicts human collaborator's hand movement. Then a robot trajectory adaptive planning submodule is developed that takes the noisy movement prediction signal into consideration for optimization. To validate the proposed systems, we first collect a new human manipulation dataset that can supplement the previous publicly available dataset with motion capture data to serve as the ground truth of hand location. We then integrate the algorithm with a six degree-of-freedom robot manipulator that can collaborate with human workers on a set of trained manipulation actions, and it is shown that such a robot system outperforms the one without movement prediction in terms of collision avoidance. We verify the effectiveness of the proposed motion prediction and robot trajectory planning approaches in both simulated and physical experiments. To the best of the authors' knowledge, it is the first time that a deep model based movement prediction system is utilized and is proven effective in human-robot collaboration scenario for enhanced safety.
{"title":"Collision-free trajectory planning in human-robot interaction through hand movement prediction from vision","authors":"Yiwei Wang, Xin Ye, Yezhou Yang, Wenlong Zhang","doi":"10.1109/HUMANOIDS.2017.8246890","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246890","url":null,"abstract":"We present a framework from vision based hand movement prediction in a real-world human-robot collaborative scenario for safety guarantee. We first propose a perception submodule that takes in visual data solely and predicts human collaborator's hand movement. Then a robot trajectory adaptive planning submodule is developed that takes the noisy movement prediction signal into consideration for optimization. To validate the proposed systems, we first collect a new human manipulation dataset that can supplement the previous publicly available dataset with motion capture data to serve as the ground truth of hand location. We then integrate the algorithm with a six degree-of-freedom robot manipulator that can collaborate with human workers on a set of trained manipulation actions, and it is shown that such a robot system outperforms the one without movement prediction in terms of collision avoidance. We verify the effectiveness of the proposed motion prediction and robot trajectory planning approaches in both simulated and physical experiments. To the best of the authors' knowledge, it is the first time that a deep model based movement prediction system is utilized and is proven effective in human-robot collaboration scenario for enhanced safety.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121171970","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.2017.8246975
Kunpeng Yao, Mohsen Kaboli, G. Cheng
In robotic tasks, object recognition and discrimination can be realized according to their physical properties, such as color, shape, stiffness, and surface textures. However, these external properties may fail if they are similar or even identical. In this case, internal properties of the objects can be considered, for example, the center of mass. Center of mass is an important inherent physical property of objects; however, due to the difficulties in its determination, it has never been applied in object discrimination tasks. In this work, we present a tactile-based approach to explore the center of mass of rigid objects and apply it in robotic object discrimination tasks. This work comprises three aspects: (a) continuous estimation of the target object's geometric information, (b) exploration of the center of mass, and (c) object discrimination based on the center of mass features. Experimental results show that by following our proposed approach, the center of mass of experimental objects can be accurately estimated, and objects of identical external properties but different mass distributions can be successfully discriminated. Our approach is also robust against the textural properties and stiffness of experimental objects.
{"title":"Tactile-based object center of mass exploration and discrimination","authors":"Kunpeng Yao, Mohsen Kaboli, G. Cheng","doi":"10.1109/HUMANOIDS.2017.8246975","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246975","url":null,"abstract":"In robotic tasks, object recognition and discrimination can be realized according to their physical properties, such as color, shape, stiffness, and surface textures. However, these external properties may fail if they are similar or even identical. In this case, internal properties of the objects can be considered, for example, the center of mass. Center of mass is an important inherent physical property of objects; however, due to the difficulties in its determination, it has never been applied in object discrimination tasks. In this work, we present a tactile-based approach to explore the center of mass of rigid objects and apply it in robotic object discrimination tasks. This work comprises three aspects: (a) continuous estimation of the target object's geometric information, (b) exploration of the center of mass, and (c) object discrimination based on the center of mass features. Experimental results show that by following our proposed approach, the center of mass of experimental objects can be accurately estimated, and objects of identical external properties but different mass distributions can be successfully discriminated. Our approach is also robust against the textural properties and stiffness of experimental objects.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125804649","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.2017.8246883
A. Sarmadi, Maziar Ahmad Sharbafi, A. Seyfarth
Human body as a segmented inverted pendulum is an unstable system needing posture control for balancing. Using the inverted pendulum as a template representing human balance in quiet standing, neuromuscular models can be employed for understanding posture control. Feedback control is implemented in human neuromuscular systems by reflex signals. In this paper, we want to realize which type of reflex signals are the most advantageous ones in posture control. As common reflex signals, muscle length, velocity and force are examined. Simulations, stability and robustness analyses show that combination of force and velocity reflexes results in a stable system with the largest basin of attraction, the most robustness against perturbations and the best performance. In addition, a proposed model with two antagonistic muscle can explain human moderate oscillations at quiet standing, using the metabolic effort under optimal control argumentation.
{"title":"Reflex control of body posture in standing","authors":"A. Sarmadi, Maziar Ahmad Sharbafi, A. Seyfarth","doi":"10.1109/HUMANOIDS.2017.8246883","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246883","url":null,"abstract":"Human body as a segmented inverted pendulum is an unstable system needing posture control for balancing. Using the inverted pendulum as a template representing human balance in quiet standing, neuromuscular models can be employed for understanding posture control. Feedback control is implemented in human neuromuscular systems by reflex signals. In this paper, we want to realize which type of reflex signals are the most advantageous ones in posture control. As common reflex signals, muscle length, velocity and force are examined. Simulations, stability and robustness analyses show that combination of force and velocity reflexes results in a stable system with the largest basin of attraction, the most robustness against perturbations and the best performance. In addition, a proposed model with two antagonistic muscle can explain human moderate oscillations at quiet standing, using the metabolic effort under optimal control argumentation.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116270909","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.2017.8246878
Iretiayo Akinola, Boyuan Chen, Jonathan Koss, Aalhad Patankar, Jacob Varley, P. Allen
This paper describes a novel hierarchical system for shared control of a humanoid robot. Our framework uses a low-bandwidth Brain Computer Interface (BCI) to interpret electroencephalography (EEG) signals via Steady-State Visual Evoked Potentials (SSVEP). This BCI allows a user to reliably interact with the humanoid. Our system clearly delineates between autonomous robot operation and human-guided intervention and control. Our shared-control system leverages the ability of the robot to accomplish low level tasks on its own, while the user assists the robot with high level directions when needed. This partnership prevents fatigue of the human controller by not requiring continuous BCI control to accomplish tasks which can be automated. We have tested the system in simulation and in real physical settings with multiple subjects using a Fetch mobile manipulator. Working together, the robot and human controller were able to accomplish tasks such as navigation, pick and place, and table clean up.
{"title":"Task level hierarchical system for BCI-enabled shared autonomy","authors":"Iretiayo Akinola, Boyuan Chen, Jonathan Koss, Aalhad Patankar, Jacob Varley, P. Allen","doi":"10.1109/HUMANOIDS.2017.8246878","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246878","url":null,"abstract":"This paper describes a novel hierarchical system for shared control of a humanoid robot. Our framework uses a low-bandwidth Brain Computer Interface (BCI) to interpret electroencephalography (EEG) signals via Steady-State Visual Evoked Potentials (SSVEP). This BCI allows a user to reliably interact with the humanoid. Our system clearly delineates between autonomous robot operation and human-guided intervention and control. Our shared-control system leverages the ability of the robot to accomplish low level tasks on its own, while the user assists the robot with high level directions when needed. This partnership prevents fatigue of the human controller by not requiring continuous BCI control to accomplish tasks which can be automated. We have tested the system in simulation and in real physical settings with multiple subjects using a Fetch mobile manipulator. Working together, the robot and human controller were able to accomplish tasks such as navigation, pick and place, and table clean up.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130733259","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.2017.8246880
Yuzuko C. Nakamura, Daniel M. Troniak, Alberto Rodriguez, M. T. Mason, N. Pollard
The recent ubiquity of high-framerate (120 fps and higher) handheld cameras creates the opportunity to study human grasping at a greater level of detail than normal speed cameras allow. We first collected 91 slow-motion interactions with objects in a convenience store setting. We then annotated the actions through the lenses of various existing manipulation taxonomies. We found manipulation, particularly the process of forming a grasp, is complicated and proceeds quickly. Our dataset shows that there are many ways that people deal with clutter in order to form a strong grasp of an object. It also reveals several errors and how people recover from them. Though annotating motions in detail is time-consuming, the annotation systems we used nevertheless leave out important aspects of understanding manipulation actions, such as how the environment is functioning as a “finger” of sorts, how different parts of the hand can be involved in different grasping tasks, and high-level intent.
{"title":"The complexities of grasping in the wild","authors":"Yuzuko C. Nakamura, Daniel M. Troniak, Alberto Rodriguez, M. T. Mason, N. Pollard","doi":"10.1109/HUMANOIDS.2017.8246880","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246880","url":null,"abstract":"The recent ubiquity of high-framerate (120 fps and higher) handheld cameras creates the opportunity to study human grasping at a greater level of detail than normal speed cameras allow. We first collected 91 slow-motion interactions with objects in a convenience store setting. We then annotated the actions through the lenses of various existing manipulation taxonomies. We found manipulation, particularly the process of forming a grasp, is complicated and proceeds quickly. Our dataset shows that there are many ways that people deal with clutter in order to form a strong grasp of an object. It also reveals several errors and how people recover from them. Though annotating motions in detail is time-consuming, the annotation systems we used nevertheless leave out important aspects of understanding manipulation actions, such as how the environment is functioning as a “finger” of sorts, how different parts of the hand can be involved in different grasping tasks, and high-level intent.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125921816","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.2017.8246893
Jaesung Oh, Buyoun Cho, Jun-Ho Oh
This paper proposes a method to solve the redundancy problem by using an optimized arm angle with a 7DOF humanoid arm controlled by a 6DOF remote controller. This study presents a method to determine the feasible arm angle range within which joint limits, self-collision, and singularity do not occur, using the characteristic that the configuration of the arm changes according to the arm angle, even when the end-effector's desired task is satisfied. When the arm angle is applied, the multivariate optimization problem to solve the redundancy problem can be simply expressed as a univariate optimization problem. To verify the effectiveness of the proposed method, experiments were performed using a 6DOF data arm and the DRC-HUBO+ humanoid platform. We confirmed that the robot moves flexibly by finding the optimal arm angle so that sub-tasks such as joint constraints, selfcollision, and singularity can be satisfied while performing the desired task of the end-effector.
{"title":"Remote control for redundant humanoid arm using optimized arm angle","authors":"Jaesung Oh, Buyoun Cho, Jun-Ho Oh","doi":"10.1109/HUMANOIDS.2017.8246893","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246893","url":null,"abstract":"This paper proposes a method to solve the redundancy problem by using an optimized arm angle with a 7DOF humanoid arm controlled by a 6DOF remote controller. This study presents a method to determine the feasible arm angle range within which joint limits, self-collision, and singularity do not occur, using the characteristic that the configuration of the arm changes according to the arm angle, even when the end-effector's desired task is satisfied. When the arm angle is applied, the multivariate optimization problem to solve the redundancy problem can be simply expressed as a univariate optimization problem. To verify the effectiveness of the proposed method, experiments were performed using a 6DOF data arm and the DRC-HUBO+ humanoid platform. We confirmed that the robot moves flexibly by finding the optimal arm angle so that sub-tasks such as joint constraints, selfcollision, and singularity can be satisfied while performing the desired task of the end-effector.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121016842","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.2017.8239552
Qingbiao Li, Iordanis Chatzinikolaidis, Yiming Yang, S. Vijayakumar, Zhibin Li
This paper presents an estimation scheme to control foot placement for achieving a desired dynamic walking velocity in presence of sensor and model errors. Inevitable discrepancies, such as sensors5 noise, delay, and modelling errors, degrade the performance of model-based control methods or even cause instabilities. To resolve these issues, an on-line parameter estimation approach based on Tikhonov regularisation is formulated using measurement data, which is particularly robust for more accurately approximating the dynamics. The proposed scheme initially uses the foot placement predicted by the linear inverted pendulum model, while the control parameters are being optimised using adequate measurements to represent the real dynamics within and in-between steps; and then, the estimation based control is used to predict the future foot placement accurately in the presence of discrepancies.
{"title":"Robust foot placement control for dynamic walking using online parameter estimation","authors":"Qingbiao Li, Iordanis Chatzinikolaidis, Yiming Yang, S. Vijayakumar, Zhibin Li","doi":"10.1109/HUMANOIDS.2017.8239552","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8239552","url":null,"abstract":"This paper presents an estimation scheme to control foot placement for achieving a desired dynamic walking velocity in presence of sensor and model errors. Inevitable discrepancies, such as sensors5 noise, delay, and modelling errors, degrade the performance of model-based control methods or even cause instabilities. To resolve these issues, an on-line parameter estimation approach based on Tikhonov regularisation is formulated using measurement data, which is particularly robust for more accurately approximating the dynamics. The proposed scheme initially uses the foot placement predicted by the linear inverted pendulum model, while the control parameters are being optimised using adequate measurements to represent the real dynamics within and in-between steps; and then, the estimation based control is used to predict the future foot placement accurately in the presence of discrepancies.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116079797","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.2017.8246885
Rahul Shome, Kostas E. Bekris
Due to high-dimensionality, many motion planners for dual-arm systems follow a decoupled approach, which does not provide guarantees. Asymptotically optimal sampling-based planners provide guarantees but in practice face scalability challenges. This work improves the computational scalability of the latter methods in this domain. It builds on top of recent advances in multi-robot motion planning, which provide guarantees without having to explicitly construct a roadmap in the composite space of all robots. The proposed framework builds roadmaps for components of a humanoid robot's kinematic chain. Then, the tensor product of these component roadmaps is searched implicitly online in a way that asymptotic optimality is provided. Appropriate heuristics from the component roadmaps are utilized for discovering the solution in the composite space effectively. Evaluation on various dual-arm problems show that the method returns paths of increasing quality, has significantly reduced space requirements and improved convergence rate relative to the standard asymptotically optimal approaches.
{"title":"Improving the scalability of asymptotically optimal motion planning for humanoid dual-arm manipulators","authors":"Rahul Shome, Kostas E. Bekris","doi":"10.1109/HUMANOIDS.2017.8246885","DOIUrl":"https://doi.org/10.1109/HUMANOIDS.2017.8246885","url":null,"abstract":"Due to high-dimensionality, many motion planners for dual-arm systems follow a decoupled approach, which does not provide guarantees. Asymptotically optimal sampling-based planners provide guarantees but in practice face scalability challenges. This work improves the computational scalability of the latter methods in this domain. It builds on top of recent advances in multi-robot motion planning, which provide guarantees without having to explicitly construct a roadmap in the composite space of all robots. The proposed framework builds roadmaps for components of a humanoid robot's kinematic chain. Then, the tensor product of these component roadmaps is searched implicitly online in a way that asymptotic optimality is provided. Appropriate heuristics from the component roadmaps are utilized for discovering the solution in the composite space effectively. Evaluation on various dual-arm problems show that the method returns paths of increasing quality, has significantly reduced space requirements and improved convergence rate relative to the standard asymptotically optimal approaches.","PeriodicalId":143992,"journal":{"name":"2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115973975","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：