var bibbase_data = {"data":"\"Loading..\"\n\n
\n\n \n\n \n\n \n \n\n \n\n \n \n\n \n\n \n
\n generated by\n \n \"bibbase.org\"\n\n \n
\n \n\n
\n\n \n\n\n
\n\n Excellent! Next you can\n create a new website with this list, or\n embed it in an existing web page by copying & pasting\n any of the following snippets.\n\n
\n JavaScript\n (easiest)\n
\n \n <script src=\"https://bibbase.org/service/mendeley/fcec045e-0e8d-3ec2-90c3-ece3d1622c91?jsonp=1&jsonp=1\"></script>\n \n
\n\n PHP\n
\n \n <?php\n $contents = file_get_contents(\"https://bibbase.org/service/mendeley/fcec045e-0e8d-3ec2-90c3-ece3d1622c91?jsonp=1\");\n print_r($contents);\n ?>\n \n
\n\n iFrame\n (not recommended)\n
\n \n <iframe src=\"https://bibbase.org/service/mendeley/fcec045e-0e8d-3ec2-90c3-ece3d1622c91?jsonp=1\"></iframe>\n \n
\n\n

\n For more details see the documention.\n

\n
\n
\n\n
\n\n This is a preview! To use this list on your own web site\n or create a new web site from it,\n create a free account. The file will be added\n and you will be able to edit it in the File Manager.\n We will show you instructions once you've created your account.\n
\n\n
\n\n

To the site owner:

\n\n

Action required! Mendeley is changing its\n API. In order to keep using Mendeley with BibBase past April\n 14th, you need to:\n

    \n
  1. renew the authorization for BibBase on Mendeley, and
    2. \n
  2. update the BibBase URL\n in your page the same way you did when you initially set up\n this page.\n
    3. \n
\n

\n\n

\n \n \n Fix it now\n

\n
\n\n
\n\n\n
\n \n \n
\n
\n  \n 2020\n \n \n (6)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n Multi-expert learning of adaptive legged locomotion.\n \n \n \n\n\n \n *Yang, C.; *Yuan, K.; Zhu, Q.; Yu, W.; and Li, Z.\n\n\n \n\n\n\n Science Robotics, 5(49). 2020.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@article{\n title = {Multi-expert learning of adaptive legged locomotion},\n type = {article},\n year = {2020},\n volume = {5},\n id = {a4c415c1-d3cd-3626-8f3d-0593c4e17d09},\n created = {2021-04-21T21:00:15.692Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:21:51.691Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {Achieving versatile robot locomotion requires motor skills that can adapt to previously unseen situations. We propose a multi-expert learning architecture (MELA) that learns to generate adaptive skills from a group of representative expert skills. During training, MELA is first initialized by a distinct set of pretrained experts, each in a separate deep neural network (DNN). Then, by learning the combination of these DNNs using a gating neural network (GNN), MELA can acquire more specialized experts and transitional skills across various locomotion modes. During runtime, MELA constantly blends multiple DNNs and dynamically synthesizes a new DNN to produce adaptive behaviors in response to changing situations. This approach leverages the advantages of trained expert skills and the fast online synthesis of adaptive policies to generate responsive motor skills during the changing tasks. Using one unified MELA framework, we demonstrated successful multiskill locomotion on a real quadruped robot that performed coherent trotting, steering, and fall recovery autonomously and showed the merit of multi-expert learning generating behaviors that can adapt to unseen scenarios.},\n bibtype = {article},\n author = {*Yang, C. and *Yuan, K. and Zhu, Q. and Yu, W. and Li, Z.},\n doi = {10.1126/scirobotics.abb2174},\n journal = {Science Robotics},\n number = {49}\n}
\n
\n\n\n
\n Achieving versatile robot locomotion requires motor skills that can adapt to previously unseen situations. We propose a multi-expert learning architecture (MELA) that learns to generate adaptive skills from a group of representative expert skills. During training, MELA is first initialized by a distinct set of pretrained experts, each in a separate deep neural network (DNN). Then, by learning the combination of these DNNs using a gating neural network (GNN), MELA can acquire more specialized experts and transitional skills across various locomotion modes. During runtime, MELA constantly blends multiple DNNs and dynamically synthesizes a new DNN to produce adaptive behaviors in response to changing situations. This approach leverages the advantages of trained expert skills and the fast online synthesis of adaptive policies to generate responsive motor skills during the changing tasks. Using one unified MELA framework, we demonstrated successful multiskill locomotion on a real quadruped robot that performed coherent trotting, steering, and fall recovery autonomously and showed the merit of multi-expert learning generating behaviors that can adapt to unseen scenarios.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Learning Pregrasp Manipulation of Objects from Ungraspable Poses.\n \n \n \n\n\n \n Sun, Z.; Yuan, K.; Hu, W.; Yang, C.; and Li, Z.\n\n\n \n\n\n\n In Proceedings - IEEE International Conference on Robotics and Automation, 2020. \n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{\n title = {Learning Pregrasp Manipulation of Objects from Ungraspable Poses},\n type = {inproceedings},\n year = {2020},\n id = {8dad914e-a1ba-3b48-b6ee-ac8a3e3588ba},\n created = {2021-04-21T21:00:15.803Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.803Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {In robotic grasping, objects are often occluded in ungraspable configurations such that no feasible grasp pose can be found, e.g. large flat boxes on the table that can only be grasped once lifted. Inspired by human bimanual manipulation, e.g. one hand to lift up things and the other to grasp, we address this type of problems by introducing pregrasp manipulation - push and lift actions. We propose a model-free Deep Reinforcement Learning framework to train feedback control policies that utilize visual information and proprioceptive states of the robot to autonomously discover robust pregrasp manipulation. The robot arm learns to push the object first towards a support surface and then lift up one side of the object, creating an object-table clearance for possible grasping solutions. Furthermore, we show the robustness of the proposed learning framework in training pregrasp policies that can be directly transferred to a real robot. Lastly, we evaluate the effectiveness and generalization ability of the learned policy in real-world experiments, and demonstrate pregrasp manipulation of objects with various sizes, shapes, weights, and surface friction.},\n bibtype = {inproceedings},\n author = {Sun, Z. and Yuan, K. and Hu, W. and Yang, C. and Li, Z.},\n doi = {10.1109/ICRA40945.2020.9196982},\n booktitle = {Proceedings - IEEE International Conference on Robotics and Automation}\n}
\n
\n\n\n
\n In robotic grasping, objects are often occluded in ungraspable configurations such that no feasible grasp pose can be found, e.g. large flat boxes on the table that can only be grasped once lifted. Inspired by human bimanual manipulation, e.g. one hand to lift up things and the other to grasp, we address this type of problems by introducing pregrasp manipulation - push and lift actions. We propose a model-free Deep Reinforcement Learning framework to train feedback control policies that utilize visual information and proprioceptive states of the robot to autonomously discover robust pregrasp manipulation. The robot arm learns to push the object first towards a support surface and then lift up one side of the object, creating an object-table clearance for possible grasping solutions. Furthermore, we show the robustness of the proposed learning framework in training pregrasp policies that can be directly transferred to a real robot. Lastly, we evaluate the effectiveness and generalization ability of the learned policy in real-world experiments, and demonstrate pregrasp manipulation of objects with various sizes, shapes, weights, and surface friction.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Decoding motor skills of artificial intelligence and human policies: A study on humanoid and human balance control.\n \n \n \n\n\n \n Yuan, K.; McGreavy, C.; Yang, C.; Wolfslag, W.; and Li, Z.\n\n\n \n\n\n\n IEEE Robotics and Automation Magazine, 27(2). 2020.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@article{\n title = {Decoding motor skills of artificial intelligence and human policies: A study on humanoid and human balance control},\n type = {article},\n year = {2020},\n volume = {27},\n id = {52857a8d-406d-323f-b5e4-7523cdaaaa19},\n created = {2021-04-21T21:00:15.846Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.846Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {From the advancement in computers, computeraided design has emerged for mechanical and electronic engineering, architecture, and many other engineering fields. Foreseeing a similar development curve and technology wave, we forecast a new emerging discipline in the near future that uses learning-aided approaches to catalyze control development, alongside other similar applications such as medicine discovery. In this article, we propose a new paradigm for using machine learning to facilitate quicker, more efficient, and more effective control development, as an alternative way of leveraging the power of machine learning in addition to other options that intend to use learning directly in real-world applications.},\n bibtype = {article},\n author = {Yuan, K. and McGreavy, C. and Yang, C. and Wolfslag, W. and Li, Z.},\n doi = {10.1109/MRA.2020.2980547},\n journal = {IEEE Robotics and Automation Magazine},\n number = {2}\n}
\n
\n\n\n
\n From the advancement in computers, computeraided design has emerged for mechanical and electronic engineering, architecture, and many other engineering fields. Foreseeing a similar development curve and technology wave, we forecast a new emerging discipline in the near future that uses learning-aided approaches to catalyze control development, alongside other similar applications such as medicine discovery. In this article, we propose a new paradigm for using machine learning to facilitate quicker, more efficient, and more effective control development, as an alternative way of leveraging the power of machine learning in addition to other options that intend to use learning directly in real-world applications.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Learning Natural Locomotion Behaviors for Humanoid Robots Using Human Bias.\n \n \n \n\n\n \n Yang, C.; Yuan, K.; Heng, S.; Komura, T.; and Li, Z.\n\n\n \n\n\n\n IEEE Robotics and Automation Letters, 5(2). 2020.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{\n title = {Learning Natural Locomotion Behaviors for Humanoid Robots Using Human Bias},\n type = {article},\n year = {2020},\n keywords = {Deep learning in robotics and automation,humanoid and bipedal locomotion,learning from demonstration},\n volume = {5},\n id = {6030ffc5-df27-3817-8091-3460b321209a},\n created = {2021-04-21T21:00:15.861Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.861Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {This letter presents a new learning framework that leverages the knowledge from imitation learning, deep reinforcement learning, and control theories to achieve human-style locomotion that is natural, dynamic, and robust for humanoids. We proposed novel approaches to introduce human bias, i.e. motion capture data and a special Multi-Expert network structure. We used the Multi-Expert network structure to smoothly blend behavioral features, and used the augmented reward design for the task and imitation rewards. Our reward design is composable, tunable, and explainable by using fundamental concepts from conventional humanoid control. We rigorously validated and benchmarked the learning framework which consistently produced robust locomotion behaviors in various test scenarios. Further, we demonstrated the capability of learning robust and versatile policies in the presence of disturbances, such as terrain irregularities and external pushes.},\n bibtype = {article},\n author = {Yang, C. and Yuan, K. and Heng, S. and Komura, T. and Li, Z.},\n doi = {10.1109/LRA.2020.2972879},\n journal = {IEEE Robotics and Automation Letters},\n number = {2}\n}
\n
\n\n\n
\n This letter presents a new learning framework that leverages the knowledge from imitation learning, deep reinforcement learning, and control theories to achieve human-style locomotion that is natural, dynamic, and robust for humanoids. We proposed novel approaches to introduce human bias, i.e. motion capture data and a special Multi-Expert network structure. We used the Multi-Expert network structure to smoothly blend behavioral features, and used the augmented reward design for the task and imitation rewards. Our reward design is composable, tunable, and explainable by using fundamental concepts from conventional humanoid control. We rigorously validated and benchmarked the learning framework which consistently produced robust locomotion behaviors in various test scenarios. Further, we demonstrated the capability of learning robust and versatile policies in the presence of disturbances, such as terrain irregularities and external pushes.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Force-Guided High-Precision Grasping Control of Fragile and Deformable Objects Using sEMG-Based Force Prediction.\n \n \n \n\n\n \n Wen, R.; Yuan, K.; Wang, Q.; Heng, S.; and Li, Z.\n\n\n \n\n\n\n IEEE Robotics and Automation Letters, 5(2). 2020.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{\n title = {Force-Guided High-Precision Grasping Control of Fragile and Deformable Objects Using sEMG-Based Force Prediction},\n type = {article},\n year = {2020},\n keywords = {Dexterous manipulation,force control,grasping},\n volume = {5},\n id = {de3ec740-3d5c-3b8e-b515-df6a83347c0b},\n created = {2021-04-21T21:00:15.895Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.895Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {Regulating contact forces with high precision is crucial for grasping and manipulating fragile or deformable objects. We aim to utilize the dexterity of human hands to regulate the contact forces for robotic hands and exploit human sensory-motor synergies in a wearable and non-invasive way. We extracted force information from the electric activities of skeletal muscles during their voluntary contractions through surface electromyography (sEMG). We built a regression model based on a Neural Network to predict the gripping force from the preprocessed sEMG signals and achieved high accuracy ($R^2 = 0.982$). Based on the force command predicted from human muscles, we developed a force-guided control framework, where force control was realized via an admittance controller that tracked the predicted gripping force reference to grasp delicate and deformable objects. We demonstrated the effectiveness of the proposed method on a set of representative fragile and deformable objects from daily life, all of which were successfully grasped without any damage or deformation.},\n bibtype = {article},\n author = {Wen, R. and Yuan, K. and Wang, Q. and Heng, S. and Li, Z.},\n doi = {10.1109/LRA.2020.2974439},\n journal = {IEEE Robotics and Automation Letters},\n number = {2}\n}
\n
\n\n\n
\n Regulating contact forces with high precision is crucial for grasping and manipulating fragile or deformable objects. We aim to utilize the dexterity of human hands to regulate the contact forces for robotic hands and exploit human sensory-motor synergies in a wearable and non-invasive way. We extracted force information from the electric activities of skeletal muscles during their voluntary contractions through surface electromyography (sEMG). We built a regression model based on a Neural Network to predict the gripping force from the preprocessed sEMG signals and achieved high accuracy ($R^2 = 0.982$). Based on the force command predicted from human muscles, we developed a force-guided control framework, where force control was realized via an admittance controller that tracked the predicted gripping force reference to grasp delicate and deformable objects. We demonstrated the effectiveness of the proposed method on a set of representative fragile and deformable objects from daily life, all of which were successfully grasped without any damage or deformation.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Unified Push Recovery Fundamentals: Inspiration from Human Study.\n \n \n \n\n\n \n McGreavy, C.; Yuan, K.; Gordon, D.; Tan, K.; Wolfslag, W.; Vijayakumar, S.; and Li, Z.\n\n\n \n\n\n\n In Proceedings - IEEE International Conference on Robotics and Automation, 2020. \n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{\n title = {Unified Push Recovery Fundamentals: Inspiration from Human Study},\n type = {inproceedings},\n year = {2020},\n id = {2fe63592-7a30-34ee-8909-f9cf65efaf09},\n created = {2021-04-21T21:00:16.142Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:16.142Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {Currently for balance recovery, humans outperform humanoid robots which use hand-designed controllers in terms of the diverse actions. This study aims to close this gap by finding core control principles that are shared across ankle, hip, toe and stepping strategies by formulating experiments to test human balance recoveries and define criteria to quantify the strategy in use. To reveal fundamental principles of balance strategies, our study shows that a minimum jerk controller can accurately replicate comparable human behaviour at the Centre of Mass level. Therefore, we formulate a general Model-Predictive Control (MPC) framework to produce recovery motions in any system, including legged machines, where the framework parameters are tuned for time-optimal performance in robotic systems.},\n bibtype = {inproceedings},\n author = {McGreavy, C. and Yuan, K. and Gordon, D. and Tan, K. and Wolfslag, W.J. and Vijayakumar, S. and Li, Z.},\n doi = {10.1109/ICRA40945.2020.9196911},\n booktitle = {Proceedings - IEEE International Conference on Robotics and Automation}\n}
\n
\n\n\n
\n Currently for balance recovery, humans outperform humanoid robots which use hand-designed controllers in terms of the diverse actions. This study aims to close this gap by finding core control principles that are shared across ankle, hip, toe and stepping strategies by formulating experiments to test human balance recoveries and define criteria to quantify the strategy in use. To reveal fundamental principles of balance strategies, our study shows that a minimum jerk controller can accurately replicate comparable human behaviour at the Centre of Mass level. Therefore, we formulate a general Model-Predictive Control (MPC) framework to produce recovery motions in any system, including legged machines, where the framework parameters are tuned for time-optimal performance in robotic systems.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2019\n \n \n (2)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n Bayesian Optimization for Whole-Body Control of High-Degree-of-Freedom Robots Through Reduction of Dimensionality.\n \n \n \n\n\n \n Yuan, K.; Chatzinikolaidis, I.; and Li, Z.\n\n\n \n\n\n\n IEEE Robotics and Automation Letters, 4(3). 2019.\n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n \n \n \n \n \n \n \n \n\n\n\n
\n 
@article{\n title = {Bayesian Optimization for Whole-Body Control of High-Degree-of-Freedom Robots Through Reduction of Dimensionality},\n type = {article},\n year = {2019},\n keywords = {Optimization and optimal control,humanoid and bipedal locomotion,humanoid robots,legged robots},\n volume = {4},\n id = {5df89826-3412-3eac-953e-6315a585a899},\n created = {2021-04-21T21:00:15.548Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.548Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {This letter aims to achieve automatic tuning of optimal parameters for whole-body control algorithms to achieve the best performance of high-DoF robots. Typically, the control parameters at a scale up to hundreds are often hand-tuned yielding sub-optimal performance. Bayesian optimization (BO) can be an option to automatically find optimal parameters. However, for high-dimensional problems, BO is often infeasible in realistic settings as we studied in this letter. Moreover, the data is too little to perform dimensionality reduction techniques, such as principal component analysis or partial least square. We hereby propose an alternating BO algorithm that iteratively learns the parameters of sub-spaces from the whole high-dimensional parametric space through interactive trials, resulting in sample efficiency and fast convergence. Furthermore, for the balancing and locomotion control of humanoids, we developed techniques of dimensionality reduction combined with the proposed ABO approach that demonstrated optimal parameters for robust whole-body control.},\n bibtype = {article},\n author = {Yuan, K. and Chatzinikolaidis, I. and Li, Z.},\n doi = {10.1109/LRA.2019.2901308},\n journal = {IEEE Robotics and Automation Letters},\n number = {3}\n}
\n
\n\n\n
\n This letter aims to achieve automatic tuning of optimal parameters for whole-body control algorithms to achieve the best performance of high-DoF robots. Typically, the control parameters at a scale up to hundreds are often hand-tuned yielding sub-optimal performance. Bayesian optimization (BO) can be an option to automatically find optimal parameters. However, for high-dimensional problems, BO is often infeasible in realistic settings as we studied in this letter. Moreover, the data is too little to perform dimensionality reduction techniques, such as principal component analysis or partial least square. We hereby propose an alternating BO algorithm that iteratively learns the parameters of sub-spaces from the whole high-dimensional parametric space through interactive trials, resulting in sample efficiency and fast convergence. Furthermore, for the balancing and locomotion control of humanoids, we developed techniques of dimensionality reduction combined with the proposed ABO approach that demonstrated optimal parameters for robust whole-body control.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Learning Whole-Body Motor Skills for Humanoids.\n \n \n \n\n\n \n Yang, C.; Yuan, K.; Merkt, W.; Komura, T.; Vijayakumar, S.; and Li, Z.\n\n\n \n\n\n\n In IEEE-RAS International Conference on Humanoid Robots, volume 2018-Novem, 2019. \n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{\n title = {Learning Whole-Body Motor Skills for Humanoids},\n type = {inproceedings},\n year = {2019},\n volume = {2018-Novem},\n id = {3631b402-eeff-31d7-88c4-75e92b58fd95},\n created = {2021-04-21T21:00:16.118Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:16.118Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {This paper presents a hierarchical framework for Deep Reinforcement Learning that acquires motor skills for a variety of push recovery and balancing behaviors, i.e., ankle, hip, foot tilting, and stepping strategies. The policy is trained in a physics simulator with realistic setting of robot model and low-level impedance control that are easy to transfer the learned skills to real robots. The advantage over traditional methods is the integration of high-level planner and feedback control all in one single coherent policy network, which is generic for learning versatile balancing and recovery motions against unknown perturbations at arbitrary locations (e.g., legs, torso). Furthermore, the proposed framework allows the policy to be learned quickly by many state-of-the-art learning algorithms. By comparing our learned results to studies of preprogrammed, special-purpose controllers in the literature, self-learned skills are comparable in terms of disturbance rejection but with additional advantages of producing a wide range of adaptive, versatile and robust behaviors.},\n bibtype = {inproceedings},\n author = {Yang, C. and Yuan, K. and Merkt, W. and Komura, T. and Vijayakumar, S. and Li, Z.},\n doi = {10.1109/HUMANOIDS.2018.8625045},\n booktitle = {IEEE-RAS International Conference on Humanoid Robots}\n}
\n
\n\n\n
\n This paper presents a hierarchical framework for Deep Reinforcement Learning that acquires motor skills for a variety of push recovery and balancing behaviors, i.e., ankle, hip, foot tilting, and stepping strategies. The policy is trained in a physics simulator with realistic setting of robot model and low-level impedance control that are easy to transfer the learned skills to real robots. The advantage over traditional methods is the integration of high-level planner and feedback control all in one single coherent policy network, which is generic for learning versatile balancing and recovery motions against unknown perturbations at arbitrary locations (e.g., legs, torso). Furthermore, the proposed framework allows the policy to be learned quickly by many state-of-the-art learning algorithms. By comparing our learned results to studies of preprogrammed, special-purpose controllers in the literature, self-learned skills are comparable in terms of disturbance rejection but with additional advantages of producing a wide range of adaptive, versatile and robust behaviors.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n
\n
\n  \n 2018\n \n \n (2)\n \n \n
\n
\n \n \n
\n \n\n \n \n \n \n \n An Improved Formulation for Model Predictive Control of Legged Robots for Gait Planning and Feedback Control.\n \n \n \n\n\n \n Yuan, K.; and Li, Z.\n\n\n \n\n\n\n In IEEE International Conference on Intelligent Robots and Systems, 2018. \n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{\n title = {An Improved Formulation for Model Predictive Control of Legged Robots for Gait Planning and Feedback Control},\n type = {inproceedings},\n year = {2018},\n id = {1c41854e-5cde-3879-bf3e-b53fca77d7eb},\n created = {2021-04-21T21:00:15.546Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.546Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {Predictive control methods for walking commonly use low dimensional models, such as a Linear Inverted Pendulum Model (LIPM), for simplifying the complex dynamics of legged robots. This paper identifies the physical limitations of the modeling methods that do not account for external disturbances, and then analyzes the issues of numerical stability of Model Predictive Control (MPC)using different models with variable receding horizons. We propose a new modeling formulation that can be used for both gait planning and feedback control in an MPC scheme. The advantages are the improved numerical stability for long prediction horizons and the robustness against various disturbances. Benchmarks were rigorously studied to compare the proposed MPC scheme with the existing ones in terms of numerical stability and disturbance rejection. The effectiveness of the controller is demonstrated in both MATLAB and Gazebo simulations.},\n bibtype = {inproceedings},\n author = {Yuan, K. and Li, Z.},\n doi = {10.1109/IROS.2018.8594309},\n booktitle = {IEEE International Conference on Intelligent Robots and Systems}\n}
\n
\n\n\n
\n Predictive control methods for walking commonly use low dimensional models, such as a Linear Inverted Pendulum Model (LIPM), for simplifying the complex dynamics of legged robots. This paper identifies the physical limitations of the modeling methods that do not account for external disturbances, and then analyzes the issues of numerical stability of Model Predictive Control (MPC)using different models with variable receding horizons. We propose a new modeling formulation that can be used for both gait planning and feedback control in an MPC scheme. The advantages are the improved numerical stability for long prediction horizons and the robustness against various disturbances. Benchmarks were rigorously studied to compare the proposed MPC scheme with the existing ones in terms of numerical stability and disturbance rejection. The effectiveness of the controller is demonstrated in both MATLAB and Gazebo simulations.\n
\n\n\n
\n\n\n
\n \n\n \n \n \n \n \n Comparison study of nonlinear optimization of step durations and foot placement for dynamic walking.\n \n \n \n\n\n \n Hu, W.; Chatzinikolaidis, I.; Yuan, K.; and Li, Z.\n\n\n \n\n\n\n In Proceedings - IEEE International Conference on Robotics and Automation, 2018. \n \n\n\n\n
\n\n\n\n \n\n \n \n doi\n  \n \n\n \n link\n  \n \n\n bibtex\n \n\n \n  \n \n abstract \n \n\n \n  \n \n 1 download\n \n \n\n \n \n \n \n \n \n \n\n  \n \n \n\n\n\n
\n 
@inproceedings{\n title = {Comparison study of nonlinear optimization of step durations and foot placement for dynamic walking},\n type = {inproceedings},\n year = {2018},\n id = {f7387ee1-0beb-3ade-9351-ec41a6e7748d},\n created = {2021-04-21T21:00:15.642Z},\n file_attached = {false},\n profile_id = {fcec045e-0e8d-3ec2-90c3-ece3d1622c91},\n last_modified = {2021-04-21T21:00:15.642Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {true},\n abstract = {This paper studies bipedal locomotion as a nonlinear optimization problem based on continuous and discrete dynamics, by simultaneously optimizing the remaining step duration, the next step duration and the foot location to achieve robustness. The linear inverted pendulum as the motion model captures the center of mass dynamics and its low-dimensionality makes the problem more tractable. We first formulate a holistic approach to search for optimality in the three-dimensional parametric space and use these results as baseline. To further improve computational efficiency, our study investigates a sequential approach with two stages of customized optimization that first optimizes the current step duration, and subsequently the duration and location of the next step. The effectiveness of both approaches is successfully demonstrated in simulation by applying different perturbations. The comparison study shows that these two approaches find mostly the same optimal solutions, but the latter requires considerably less computational time, which suggests that the proposed sequential approach is well suited for real-time implementation with a minor trade-off in optimality.},\n bibtype = {inproceedings},\n author = {Hu, W. and Chatzinikolaidis, I. and Yuan, K. and Li, Z.},\n doi = {10.1109/ICRA.2018.8461101},\n booktitle = {Proceedings - IEEE International Conference on Robotics and Automation}\n}
\n
\n\n\n
\n This paper studies bipedal locomotion as a nonlinear optimization problem based on continuous and discrete dynamics, by simultaneously optimizing the remaining step duration, the next step duration and the foot location to achieve robustness. The linear inverted pendulum as the motion model captures the center of mass dynamics and its low-dimensionality makes the problem more tractable. We first formulate a holistic approach to search for optimality in the three-dimensional parametric space and use these results as baseline. To further improve computational efficiency, our study investigates a sequential approach with two stages of customized optimization that first optimizes the current step duration, and subsequently the duration and location of the next step. The effectiveness of both approaches is successfully demonstrated in simulation by applying different perturbations. The comparison study shows that these two approaches find mostly the same optimal solutions, but the latter requires considerably less computational time, which suggests that the proposed sequential approach is well suited for real-time implementation with a minor trade-off in optimality.\n
\n\n\n
\n\n\n\n\n\n
\n
\n\n\n\n\n
\n\n\n \n\n \n \n \n \n\n
\n"}; document.write(bibbase_data.data);