Multifingered Grasping Based on Multimodal Reinforcement Learning

Liang, Hongzhuo; Cong, Lin; Hendrich, Norman; Li, Shuang; Sun, Fuchun; Zhang, Jianwei

doi:10.1109/lra.2021.3138545

Cited by 28 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, they are difficult to apply since they are constrained to the hardware considered for training and they do not take into account the hand-object interactions involved during grasp execution. The approach described in [21] deals with the high number of DoFs in the Shadow hand with a PCA-based hand synergy. Then, similarly to our proposed approach, it trains a DRL policy to grasp an object starting from a grasp pose given by an external algorithm.…”

Section: A Multi-fingered Graspingmentioning

confidence: 99%

Fast Object Segmentation Learning with Kernel-based Methods for Robotics

Ceola

Maiettini

Pasquale

et al. 2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Multi-fingered robotic hands could enable robots to perform sophisticated manipulation tasks. However, teaching a robot to grasp objects with an anthropomorphic hand is an arduous problem due to the high dimensionality of state and action spaces. Deep Reinforcement Learning (DRL) offers techniques to design control policies for this kind of problems without explicit environment or hand modeling. However, training these policies with state-of-the-art model-free algorithms is greatly challenging for multi-fingered hands. The main problem is that an efficient exploration of the environment is not possible for such high-dimensional problems, thus causing issues in the initial phases of policy optimization. One possibility to address this is to rely on off-line task demonstrations. However, oftentimes this is incredibly demanding in terms of time and computational resources.In this work, we overcome these requirements and propose the A Grasp Pose is All You Need (G-PAYN) method for the anthropomorphic hand of the iCub humanoid. We develop an approach to automatically collect task demonstrations to initialize the training of the policy. The proposed grasping pipeline starts from a grasp pose generated by an external algorithm, used to initiate the movement. Then a control policy (previously trained with the proposed G-PAYN) is used to reach and grab the object. We deployed the iCub into the MuJoCo simulator and use it to test our approach with objects from the YCB-Video dataset. The results show that G-PAYN outperforms current DRL techniques in the considered setting, in terms of success rate and execution time with respect to the baselines. The code to reproduce the experiments will be released upon acceptance.

show abstract

Section: A Multi-fingered Graspingmentioning

confidence: 99%

Fast Object Segmentation Learning with Kernel-based Methods for Robotics

Ceola

Maiettini

Pasquale

et al. 2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“… 163 To learn the finger motions and to determine when to lift an object, an algorithm was trained by combing fingertip tactile sensing, joint torques, and proprioception as well as a multimodal agent through reinforcement learning. 164 As for the aforementioned haptic feedback, it has been investigated by many researchers. This technology could not only help surgeons minimize potential side effects but also reduce errors in remote operations, tissue damage, and operation time.…”

Section: Motion Control Technologymentioning

confidence: 99%

Review of robotic systems for thoracoabdominal puncture interventional surgery

Wang,

Guo,

Zhu

et al. 2024

APL Bioengineering

View full text Add to dashboard Cite

Cancer, with high morbidity and high mortality, is one of the major burdens threatening human health globally. Intervention procedures via percutaneous puncture have been widely used by physicians due to its minimally invasive surgical approach. However, traditional manual puncture intervention depends on personal experience and faces challenges in terms of precisely puncture, learning-curve, safety and efficacy. The development of puncture interventional surgery robotic (PISR) systems could alleviate the aforementioned problems to a certain extent. This paper attempts to review the current status and prospective of PISR systems for thoracic and abdominal application. In this review, the key technologies related to the robotics, including spatial registration, positioning navigation, puncture guidance feedback, respiratory motion compensation, and motion control, are discussed in detail.

show abstract

“…Dexterous Robotic Grasping There have been works in different directions attempting dexterous grasping. Liang et al [13] map low-DOF end-effectors to high-DOF ones to solve the high-DOF grasping problem. Shao [7] models the robotic hand and object and uses optimization methods to find force closure and contact points.…”

Section: Related Workmentioning

confidence: 99%

DexRepNet: Learning Dexterous Robotic Grasping Network with Geometric and Spatial Hand-Object Representations

Liu¹,

Chen²,

Sun³

et al. 2023

Preprint

View full text Add to dashboard Cite

Robotic dexterous grasping is a challenging problem due to the high degree of freedom (DoF) and complex contacts of multi-fingered robotic hands. Existing deep reinforcement learning (DRL) based methods leverage human demonstrations to reduce sample complexity due to the high dimensional action space with dexterous grasping. However, less attention has been paid to hand-object interaction representations for high-level generalization. In this paper, we propose a novel geometric and spatial hand-object interaction representation, named DexRep, to capture dynamic object shape features and the spatial relations between hands and objects during grasping. DexRep comprises Occupancy Feature for rough shapes within sensing range by moving hands, Surface Feature for changing hand-object surface distances, and Local-Geo Feature for local geometric surface features most related to potential contacts. Based on the new representation, we propose a dexterous deep reinforcement learning method to learn a generalizable grasping policy DexRepNet. Experimental results show that our method outperforms baselines using existing representations for robotic grasping dramatically both in grasp success rate and convergence speed. It achieves a 93% grasping success rate on seen objects and higher than 80% grasping success rates on diverse objects of unseen categories in both simulation and real-world experiments.

show abstract

Multifingered Grasping Based on Multimodal Reinforcement Learning

Cited by 28 publications

References 25 publications

Fast Object Segmentation Learning with Kernel-based Methods for Robotics

Fast Object Segmentation Learning with Kernel-based Methods for Robotics

Review of robotic systems for thoracoabdominal puncture interventional surgery

DexRepNet: Learning Dexterous Robotic Grasping Network with Geometric and Spatial Hand-Object Representations

Contact Info

Product

Resources

About