Integrated linkage-driven dexterous anthropomorphic robotic hand

Kim, Uikyum; Jung, Dawoon; Jeong, Heeyoen; Park, Jongwoo; Jung, Hyun-Mok; Cheong, Joono; Choi, Hyouk Ryeol; Do, Hyunmin; Park, Chanhun

doi:10.1038/s41467-021-27261-0

Cited by 98 publications

(37 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method is based on [12] to extract the features of grippers and objects from their point clouds, such that it can train a Point Set Selection Network (PSSN), which can generate contact points on any object for any gripper. However, the method becomes data-inefficient and cumbersome when applied to multi-grasp grippers such as the Robotiq-3F gripper, the BarrettHand, and the RUTH hand [10]; and does not work on grippers with closed-loop constraints (e.g., the RUTH hand, the integrated linkage-driven hand [11]). These drawbacks are discussed in section II-C.…”

Section: B Visual Data-driven Grasping Methodsmentioning

confidence: 99%

“…However, the method becomes datainefficient and cumbersome when applied to grippers in which 'most open' and 'most closed' gripper operation modes are ambiguous, as in the Robotiq-3F gripper, the BarrettHand, or the RUTH hand [10]. Moreover, UniGrasp cannot work on grippers with closed-loop constraints (e.g., the RUTH hand, the integrated linkage-driven hand [11]) since a URDF cannot represent mechanisms with this characteristic.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

EfficientGrasp: A Unified Data-Efficient Learning to Grasp Method for Multi-Fingered Robot Hands

Baron²,

Zhang

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Autonomous grasping of novel objects that are previously unseen to a robot is an ongoing challenge in robotic manipulation. In the last decades, many approaches have been presented to address this problem for specific robot hands. The UniGrasp framework, introduced recently, has the ability to generalize to different types of robotic grippers; however, this method does not work on grippers with closed-loop constraints and is data-inefficient when applied to robot hands with multigrasp configurations. In this paper, we present EfficientGrasp, a generalized grasp synthesis and gripper control method that is independent of gripper model specifications. EfficientGrasp utilizes a gripper workspace feature rather than UniGrasp's gripper attribute inputs. This reduces memory use by 81.7% during training and makes it possible to generalize to more types of grippers, such as grippers with closed-loop constraints. The effectiveness of EfficientGrasp is evaluated by conducting object grasping experiments both in simulation and real-world; results show that the proposed method also outperforms UniGrasp when considering only grippers without closed-loop constraints. In these cases, EfficientGrasp shows 9.85% higher accuracy in generating contact points and 3.10% higher grasping success rate in simulation. The real-world experiments are conducted with a gripper with closed-loop constraints, which UniGrasp fails to handle while EfficientGrasp achieves a success rate of 83.3%. The main causes of grasping failures of the proposed method are analyzed, highlighting ways of enhancing grasp performance.

show abstract

Section: B Visual Data-driven Grasping Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

EfficientGrasp: A Unified Data-Efficient Learning to Grasp Method for Multi-Fingered Robot Hands

Baron²,

Zhang

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Then hand motions are reconstructed via inverse kinematics. There are more research activities on tool usage in the robotics community [Fang et al 2020b;Ke et al 2020Ke et al , 2021Kim et al 2021;Toussaint et al 2018;Wu et al 2019], although most of them turn it into a simpler problem by attaching the tools directly onto the robot arm. For example, chopsticks are attached to a robot arm to grasp objects in [Ke et al 2021], where human demonstrations were also used to help the learning of grasping policies.…”

Section: Tool Usagementioning

confidence: 99%

Learning to Use Chopsticks in Diverse Styles

Yang¹,

Yin²,

Liu³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, such dexterous manipulation remains a challenging problem because of the high dimensional actuation space and contact-rich model. Before the emergence of RL-based controllers, a large variety of manipulation tasks highly relied on accurate dynamics models and trajectory optimization methods [30,31,32]. For example, Williams et al [33] used the model predictive path integral control (MPPI) method to perform the task successfully, dexterous manipulation of a cube.…”

Section: Related Workmentioning

confidence: 99%

Towards Human-Level Bimanual Dexterous Manipulation with Reinforcement Learning

Chen¹,

Yang²,

Wu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Achieving human-level dexterity is an important open problem in robotics. However, tasks of dexterous hand manipulation, even at the baby level, are challenging to solve through reinforcement learning (RL). The difficulty lies in the high degrees of freedom and the required cooperation among heterogeneous agents (e.g., joints of fingers). In this study, we propose the Bimanual Dexterous Hands Benchmark (Bi-DexHands), a simulator that involves two dexterous hands with tens of bimanual manipulation tasks and thousands of target objects. Specifically, tasks in Bi-DexHands are designed to match different levels of human motor skills according to cognitive science literature. We built Bi-DexHands in the Issac Gym; this enables highly efficient RL training, reaching 30,000+ FPS by only one single NVIDIA RTX 3090. We provide a comprehensive benchmark for popular RL algorithms under different settings; this includes Single-agent/Multi-agent RL, Offline RL, Multi-task RL, and Meta RL. Our results show that the PPO type of on-policy algorithms can master simple manipulation tasks that are equivalent up to 48-month human babies (e.g., catching a flying object, opening a bottle), while multi-agent RL can further help to master manipulations that require skilled bimanual cooperation (e.g., lifting a pot, stacking blocks). Despite the success on each single task, when it comes to acquiring multiple manipulation skills, existing RL algorithms fail to work in most of the multi-task and the few-shot learning settings, which calls for more substantial development from the RL community.Our project is open sourced at https://github.com/PKU-MARL/DexterousHands 1 .Preprint. Under review.

show abstract

Integrated linkage-driven dexterous anthropomorphic robotic hand

Cited by 98 publications

References 44 publications

EfficientGrasp: A Unified Data-Efficient Learning to Grasp Method for Multi-Fingered Robot Hands

EfficientGrasp: A Unified Data-Efficient Learning to Grasp Method for Multi-Fingered Robot Hands

Learning to Use Chopsticks in Diverse Styles

Towards Human-Level Bimanual Dexterous Manipulation with Reinforcement Learning

Contact Info

Product

Resources

About