A Grasp Pose is All You Need: Learning Multi-fingered Grasping with Deep Reinforcement Learning from Vision and Touch
Istituto Italiano di Tecnologia (IIT)（意大利技术研究院）；Laboratory for Computational and Statistical Learning (LCSL)（计算与统计学习实验室）；Machine Learning Genoa Center (MaLGa)（热那亚机器学习中心）；University of Genoa（热那亚大学）；Center for Brains, Minds and Machines (CBMM)（大脑、心智与机器中心）；Massachusetts Institute of Technology (MIT)（麻省理工学院）
Federico Ceola, Elisa Maiettini, Lorenzo Rosasco, Lorenzo Natale
备注：
IROS 2023
发表信息：
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Detroit, MI, 2023
Multi-fingered robotic hands have potential to 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, state-of-the-art model-free algorithms have proven inefficient for learning such policies. The main problem is that the exploration of the environment is unfeasible for such high-dimensional problems, thus hampering the initial phases of policy optimization. One possibility to address this is to rely on off-line task demonstrations, but, oftentimes, this is too demanding in terms of time and computational resources. To address these problems, we 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. 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 is released together with the paper with an open source license.
University of California, Merced（加州大学默塞德分校）；National Institute of Standards and Technology (NIST)（美国国家标准与技术研究院）
This technical report presents an introduction to different aspects of multi-fingered robot grasping. After having introduced relevant mathematical background for modeling, form and force closure are discussed. Next, we present an overview of various grasp planning algorithms with the objective of illustrating different approaches to solve this problem. Finally, we discuss grasp performance benchmarking.
NVIDIA（英伟达）；Georgia Institute of Technology（佐治亚理工学院）；University of Utah（犹他大学）
Using simulation to train robot manipulation policies holds the promise of an almost unlimited amount of training data, generated safely out of harm's way. One of the key challenges of using simulation, to date, has been to bridge the reality gap, so that policies trained in simulation can be deployed in the real world. We explore the reality gap in the context of learning a contextual policy for multi-fingered robotic grasping. We propose a Grasping Objects Approach for Tactile (GOAT) robotic hands, learning to overcome the reality gap problem. In our approach we use human hand motion demonst…
RESPRECT: Speeding-up Multi-fingered Grasping with Residual Reinforcement Learning
Istituto Italiano di Tecnologia (IIT)（意大利技术研究所）；Laboratory for Computational and Statistical Learning (LCSL)（计算与统计学习实验室）；Machine Learning Genoa Center (MaLGa)（热那亚机器学习中心）；University of Genoa（热那亚大学）；Center for Brains, Minds and Machines (CBMM)（大脑、心智与机器中心）
ContactDexNet: Multi-fingered Robotic Hand Grasping in Cluttered Environments through Hand-object Contact Semantic Mapping
TAMS (Technical Aspects of Multimodal Systems), Department of Informatics, Universität Hamburg（乌尔姆大学信息学院技术多模态系统部门）；Agile Robots AG（敏捷机器人公司）；Technical University of Munich（慕尼黑技术大学）
Multi-fingered robotic grasping is an undeniable stepping stone to universal picking and dexterous manipulation. Yet, multi-fingered grippers remain challenging to control because of their rich nonsmooth contact dynamics or because of sensor noise. In this work, we aim to plan hand configurations by performing Bayesian posterior inference through the full stochastic forward simulation of the robot in its environment, hence robustly accounting for many of the uncertainties in the system. While previous methods either relied on simplified surrogates of the likelihood function or attempted to lea…
Recent advances in on-policy reinforcement learning (RL) methods enabled learning agents in virtual environments to master complex tasks with high-dimensional and continuous observation and action spaces. However, leveraging this family of algorithms in multi-fingered robotic grasping remains a challenge due to large sim-to-real fidelity gaps and the high sample complexity of on-policy RL algorithms. This work aims to bridge these gaps by first reinforcement-learning a multi-fingered robotic grasping policy in simulation that operates in the pixel space of the input: a single depth image. Usin…
In this paper we explore state-of-the-art underactuated, compliant robot gripper designs through looking at their performance on a generic grasping task. Starting from a state of the art open gripper design, we propose design modifications,and importantly, evaluate all designs on a grasping experiment involving a selection of objects resulting in 3600 object-gripper interactions. Interested in non-planned grasping but rather on a design's generic performance, we explore the influence of object shape, pose and orientation relative to the gripper and its finger number and configuration. Using op…
This letter proposes a novel single-fingered reconfigurable robotic gripper for grasping objects in narrow working spaces. The finger of the developed gripper realizes two configurations, namely, the insertion and grasping modes, using only a single motor. In the insertion mode, the finger assumes a thin shape such that it can insert its tip into a narrow space. The grasping mode of the finger is activated through a folding mechanism. Mode switching can be achieved in two ways: switching the mode actively by a motor, or combining passive rotation of the fingertip through contact with the suppo…
Soft grippers are gaining momentum across applications due to their flexibility and dexterity. However, the infinite-dimensionality and non-linearity associated with soft robots challenge modeling and closed-loop control of soft grippers to perform grasping tasks. To solve this problem, data-driven methods have been proposed. Most data-driven methods rely on intensive model learning in simulation or offline, and as such it may be hard to generalize across different settings not explicitly trained upon and in physical robot testing where online control is required. In this paper, we propose an …
University of Bristol（布里斯托大学）；University of the West of England（西英格兰大学）；Bristol Robotics Laboratory（布里斯托机器人实验室）
Tactile sensing is used by humans when grasping to prevent us dropping objects. One key facet of tactile sensing is slip detection, which allows a gripper to know when a grasp is failing and take action to prevent an object being dropped. This study demonstrates the slip detection capabilities of the recently developed Tactile Model O (T-MO) by using support vector machines to detect slip and test multiple slip scenarios including responding to the onset of slip in real time with eleven different objects in various grasps. We demonstrate the benefits of slip detection in grasping by testing tw…
Learning-based approaches to grasp planning are preferred over analytical methods due to their ability to better generalize to new, partially observed objects. However, data collection remains one of the biggest bottlenecks for grasp learning methods, particularly for multi-fingered hands. The relatively high dimensional configuration space of the hands coupled with the diversity of objects common in daily life requires a significant number of samples to produce robust and confident grasp success classifiers. In this paper, we present the first active deep learning approach to grasping that se…
University of Bristol（布里斯托大学）；Bristol Robotics Laboratory（布里斯托机器人实验室）
Bringing tactile sensation to robotic hands will allow for more effective grasping, along with the wide range of benefits of human-like touch. Here we present a 3D-printed, three-fingered tactile robot hand comprising an OpenHand Model O customized to house a TacTip soft biomimetic tactile sensor in the distal phalanx of each finger. We expect that combining the grasping capabilities of this underactuated hand with sophisticated tactile sensing will result in an effective platform for robot hand research -- the Tactile Model O (T-MO). The design uses three JeVois machine vision systems, each c…
Ecole des Ponts（巴黎路桥学院）；Univ Gustave Eiffel（古斯塔夫·埃菲尔大学）；CNRS（法国国家科学研究中心）；NAVER LABS Europe（NAVER欧洲实验室）
Robots with multi-fingered grippers could perform advanced manipulation tasks for us if we were able to properly specify to them what to do. In this study, we take a step in that direction by making a robot grasp an object like a grasping demonstration performed by a human. We propose a novel optimization-based approach for transferring human grasp demonstrations to any multi-fingered grippers, which produces robotic grasps that mimic the human hand orientation and the contact area with the object, while alleviating interpenetration. Extensive experiments with the Allegro and BarrettHand gripp…
Adaptive Motion Planning for Multi-fingered Functional Grasp via Force Feedback
School of Information and Communication and Engineering, Dalian University of Technology（大连理工大学信息与通信工程学院）；Shenyang Institute of Automation, Chinese Academy of Sciences（中国科学院沈阳自动化研究所）
Imperial College London（帝国理工学院）；Ocado Technology（奥凯多科技）
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 me…
This thesis is concerned with deriving planning algorithms for robot manipulators. Manipulation has two effects, the robot has a physical effect on the object, and it also acquires information about the object. This thesis presents algorithms that treat both problems. First, I present an extension of the well-known piano mover's problem where a robot pushing an object must plan its movements as well as those of the object. This requires simultaneous planning in the joint space of the robot and the configuration space of the object, in contrast to the original problem which only requires planni…
Sequential Multi-Object Grasping with One Dexterous Hand
Thomas Lord Department of Computer Science, University of Southern California, USA（美国南加州大学计算机科学系）；Fudan University, China（复旦大学）；INSAIT, Sofia University “St. Kliment Ohridski”, Bulgaria（保加利亚索菲亚大学“圣克莱孟·奥赫里茨基”学院）
Columbia University（哥伦比亚大学）；Robotics at Google（谷歌机器人）
Vision-based grasping systems typically adopt an open-loop execution of a planned grasp. This policy can fail due to many reasons, including ubiquitous calibration error. Recovery from a failed grasp is further complicated by visual occlusion, as the hand is usually occluding the vision sensor as it attempts another open-loop regrasp. This work presents MAT, a tactile closed-loop method capable of realizing grasps provided by a coarse initial positioning of the hand above an object. Our algorithm is a deep reinforcement learning (RL) policy optimized through the clipped surrogate objective wit…
Osaka University（大阪大学）；Meiji University（明治大学）；National Institute of Advanced Industrial Science and Technology（产业技术综合研究所）
Synergy supplies a practical approach for expressing various postures of a multi-fingered hand. However, a conventional synergy defined for reproducing grasping postures cannot perform general-purpose tasks expected for a multi-fingered hand. Locking the position of particular fingers is essential for a multi-fingered hand to manipulate an object. When using conventional synergy based control to manipulate an object, which requires locking some fingers, the coordination of joints is heavily restricted, decreasing the dexterity of the hand. We propose a functionally divided manipulation synergy…
TouchDrive: Electronics-Free Tactile Sensing Interface for Assistive Grasping
Division of Microsystems Technology, Department of Electrical Engineering（微系统技术部门）；Cyber-Physical Systems Lab, Department of Information Technology（网络物理系统实验室）
We present an attention based visual analysis framework to compute grasp-relevant information in order to guide grasp planning using a multi-fingered robotic hand. Our approach uses a computational visual attention model to locate regions of interest in a scene, and uses a deep convolutional neural network to detect grasp type and point for a sub-region of the object presented in a region of interest. We demonstrate the proposed framework in object grasping tasks, in which the information generated from the proposed framework is used as prior information to guide the grasp planning. Results sh…
We propose a novel approach to multi-fingered grasp planning leveraging learned deep neural network models. We train a voxel-based 3D convolutional neural network to predict grasp success probability as a function of both visual information of an object and grasp configuration. We can then formulate grasp planning as inferring the grasp configuration which maximizes the probability of grasp success. In addition, we learn a prior over grasp configurations as a mixture density network conditioned on our voxel-based object representation. We show that this object conditional prior improves grasp …
University of Liverpool（利物浦大学）；King’s College London（伦敦国王学院）；Monash University（莫纳什大学）
Picking up transparent objects is still a challenging task for robots. The visual properties of transparent objects such as reflection and refraction make the current grasping methods that rely on camera sensing fail to detect and localise them. However, humans can handle the transparent object well by first observing its coarse profile and then poking an area of interest to get a fine profile for grasping. Inspired by this, we propose a novel framework of vision-guided tactile poking for transparent objects grasping. In the proposed framework, a segmentation network is first used to predict t…
Dexterous robotic hands have the capability to interact with a wide variety of household objects to perform tasks like grasping. However, learning robust real world grasping policies for arbitrary objects has proven challenging due to the difficulty of generating high quality training data. In this work, we propose a learning system (ISAGrasp) for leveraging a small number of human demonstrations to bootstrap the generation of a much larger dataset containing successful grasps on a variety of novel objects. Our key insight is to use a correspondence-aware implicit generative model to deform ob…
Spirit AeroSystems（势必锐航空系统公司）；Wichita State University（威奇托州立大学）
Humans can determine a proper strategy to grasp an object according to the measured physical attributes or the prior knowledge of the object. This paper proposes an approach to determining the strategy of dexterous grasping by using an anthropomorphic robotic hand simply based on a label or a description of an object. Object attributes are parsed from natural-language descriptions and augmented with an object knowledge base that is scraped from retailer websites. A novel metric named joint probability distance is defined to measure distance between object attributes. The probability distributi…
Deep learning is an established framework for learning hierarchical data representations. While compute power is in abundance, one of the main challenges in applying this framework to robotic grasping has been obtaining the amount of data needed to learn these representations, and structuring the data to the task at hand. Among contemporary approaches in the literature, we highlight key properties that have encouraged the use of deep learning techniques, and in this paper, detail our experience in developing a simulator for collecting cylindrical precision grasps of a multi-fingered dexterous …
Embedding high-resolution touch across robotic hands enables adaptive human-like grasping
Institute for Artificial Intelligence（人工智能研究所）；Peking University（北京大学）；Beijing Institute for General Artificial Intelligence（北京通用人工智能研究院）；PKU-Wuhan Institute for Artificial Intelligence（北大武汉人工智能研究所）；College of Engineering（工程学院）；School of Engineering and Materials Science（工程与材料科学学院）
Hybrid Gripper with Passive Pneumatic Soft Joints for Grasping Deformable Thin Objects
School of Mechanical Engineering, Hanoi University of Science and Technology（河内科技大学机械工程学院）；Department of Computer Science, University of Liverpool（利物浦大学计算机科学系）；Graduate School of Information Science and Technology, The University of Tokyo（东京大学信息科学与技术研究生院）
Grasping is an incredible ability of animals using their arms and limbs in their daily life. The human hand is an especially astonishing multi-fingered tool for precise grasping, which helped humans to develop the modern world. The implementation of the human grasp to virtual reality and telerobotics is always interesting and challenging at the same time. In this work, authors surveyed, studied, and analyzed the human hand-grasping behavior for the possibilities of haptic grasping in the virtual and remote environment. This work is focused on the motion and force analysis of fingers in human h…
DexRepNet: Learning Dexterous Robotic Grasping Network with Geometric and Spatial Hand-Object Representations
College of Control Science and Engineering, Zhejiang University（浙江大学控制科学与工程学院）；Key Lab of CS&AUS of Zhejiang Province（浙江省计算机与应用系统安全重点实验室）；Department of Computer Science, National University of Singapore（新加坡国立大学计算机科学系）
DeepMind；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）；Google X Robotics（Google X 机器人）；Tandon School of Engineering, New York University（纽约大学坦登工程学院）；Stanford University（斯坦福大学）
Grasping objects under uncertainty remains an open problem in robotics research. This uncertainty is often due to noisy or partial observations of the object pose or shape. To enable a robot to react appropriately to unforeseen effects, it is crucial that it continuously takes sensor feedback into account. While visual feedback is important for inferring a grasp pose and reaching for an object, contact feedback offers valuable information during manipulation and grasp acquisition. In this paper, we use model-free deep reinforcement learning to synthesize control policies that exploit contact s…
Reliably planning fingertip grasps for multi-fingered hands lies as a key challenge for many tasks including tool use, insertion, and dexterous in-hand manipulation. This task becomes even more difficult when the robot lacks an accurate model of the object to be grasped. Tactile sensing offers a promising approach to account for uncertainties in object shape. However, current robotic hands tend to lack full tactile coverage. As such, a problem arises of how to plan and execute grasps for multi-fingered hands such that contact is made with the area covered by the tactile sensors. To address thi…
Modern approaches to grasp planning often involve deep learning. However, there are only a few large datasets of labelled grasping examples on physical robots, and available datasets involve relatively simple planar grasps with two-fingered grippers. Here we present: 1) a new human grasp demonstration method that facilitates rapid collection of naturalistic grasp examples, with full six-degree-of-freedom gripper positioning; and 2) a dataset of roughly forty thousand successful grasps on 109 different rigid objects with the RightHand Robotics three-fingered ReFlex gripper.
Beijing Institute of General Artificial Intelligence (BIGAI)（北京通用人工智能研究院）；Tsinghua University（清华大学）；Peking University（北京大学）
Generating dexterous grasping has been a long-standing and challenging robotic task. Despite recent progress, existing methods primarily suffer from two issues. First, most prior arts focus on a specific type of robot hand, lacking the generalizable capability of handling unseen ones. Second, prior arts oftentimes fail to rapidly generate diverse grasps with a high success rate. To jointly tackle these challenges with a unified solution, we propose GenDexGrasp, a novel hand-agnostic grasping algorithm for generalizable grasping. GenDexGrasp is trained on our proposed large-scale multi-hand gra…
Hunan University（湖南大学）；Bielefeld University（比勒费尔德大学）；University of the West of England（西英格兰大学）
In this work, we focus on improving the robot's dexterous capability by exploiting visual sensing and adaptive force control. TeachNet, a vision-based teleoperation learning framework, is exploited to map human hand postures to a multi-fingered robot hand. We augment TeachNet, which is originally based on an imprecise kinematic mapping and position-only servoing, with a biomimetic learning-based compliance control algorithm for dexterous manipulation tasks. This compliance controller takes the mapped robotic joint angles from TeachNet as the desired goal, computes the desired joint torques. It…
