scholarly journals Grasp Stability Prediction for a Dexterous Robotic Hand Combining Depth Vision and Haptic Bayesian Exploration

2021 ◽  
Vol 8 ◽  
Author(s):  
Muhammad Sami Siddiqui ◽  
Claudio Coppola ◽  
Gokhan Solak ◽  
Lorenzo Jamone
Keyword(s):  
Tactile Feedback ◽  
Probabilistic Methods ◽  
Robotic Manipulation ◽  
Uniform Grid ◽  
Robotic Hand ◽  
Stability Prediction ◽  
Grasp Stability ◽  
Object Pose Estimation ◽  
Depth Vision ◽  
Unknown Objects

Grasp stability prediction of unknown objects is crucial to enable autonomous robotic manipulation in an unstructured environment. Even if prior information about the object is available, real-time local exploration might be necessary to mitigate object modelling inaccuracies. This paper presents an approach to predict safe grasps of unknown objects using depth vision and a dexterous robot hand equipped with tactile feedback. Our approach does not assume any prior knowledge about the objects. First, an object pose estimation is obtained from RGB-D sensing; then, the object is explored haptically to maximise a given grasp metric. We compare two probabilistic methods (i.e. standard and unscented Bayesian Optimisation) against random exploration (i.e. uniform grid search). Our experimental results demonstrate that these probabilistic methods can provide confident predictions after a limited number of exploratory observations, and that unscented Bayesian Optimisation can find safer grasps, taking into account the uncertainty in robot sensing and grasp execution.

Visual Grasp Planning for Unknown Objects Using a Multifingered Robotic Hand

2013 ◽  
Vol 18 (3) ◽  
pp. 1050-1059 ◽  
Author(s):  
Vincenzo Lippiello ◽  
Fabio Ruggiero ◽  
Bruno Siciliano ◽  
Luigi Villani
Keyword(s):  
Robotic Hand ◽  
Grasp Planning ◽  
Unknown Objects

Design and slip prevention control of a multi-sensory anthropomorphic prosthetic hand

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Bo Zeng ◽  
Hongwei Liu ◽  
Hongzhou Song ◽  
Zhe Zhao ◽  
Shaowei Fan ◽  
...  
Keyword(s):  
Real Time ◽  
Upper Limb ◽  
Tactile Feedback ◽  
Prosthetic Hand ◽  
Discrete Wavelet ◽  
Accuracy Rate ◽  
Content Type ◽  
Grasp Stability ◽  
Grasp Force ◽  
Slip Detection

Purpose The purpose of this paper is to design a multi-sensory anthropomorphic prosthetic hand and a grasping controller that can detect the slip and automatically adjust the grasping force to prevent the slip. Design/methodology/approach To improve the dexterity, sensing, controllability and practicability of a prosthetic hand, a modular and multi-sensory prosthetic hand was presented. In addition, a slip prevention control based on the tactile feedback was proposed to improve the grasp stability. The proposed controller identifies slippages through detecting the high-frequency vibration signal at the sliding surface in real time and the discrete wavelet transform (DWT) was used to extract the eigenvalues to identify slippages. Once the slip is detected, a direct-feedback method of adjusting the grasp force related with the sliding times was used to prevent it. Furthermore, the stiffness of different objects was estimated and used to improve the grasp force control. The performances of the stiffness estimation, slip detection and slip control are experimentally evaluated. Findings It was found from the experiment of stiffness estimation that the accuracy rate of identification of the hard metal bottle could reach to 90%, while the accuracy rate of identification of the plastic bottles could reach to 80%. There was a small misjudgment rate in the identification of hard and soft plastic bottles. The stiffness of soft plastic bottles, hard plastic bottles and metal bottles were 0.64 N/mm, 1.36 N/mm and 32.55 N/mm, respectively. The results of slip detection and control show that the proposed prosthetic hand with a slip prevention controller can fast and effectively detect and prevent the slip for different disturbances, which has a certain application prospect. Practical implications Due to the small size, low weight, high integration and modularity, the prosthetic hand is easily applied to upper-limb amputees. Meanwhile, the method of the slip prevention control can be used for upper-limb amputees to complete more tasks stably in daily lives. Originality/value A multi-sensory anthropomorphic prosthetic hand is designed, and a method of stable grasps control based on slip detection by a tactile sensor on the fingertip is proposed. The method combines the stiffness estimation of the object and the real-time slip detection based on DWT with the design of the proportion differentiation robust controller based on a disturbance observer and the force controller to achieve slip prevention and stable grasps. It is verified effectively by the experiments and is easy to be applied to commercial prostheses.

scholarly journals A Differentiable Extended Kalman Filter for Object Tracking Under Sliding Regime

2021 ◽  
Vol 8 ◽  
Author(s):  
Nicola A. Piga ◽  
Ugo Pattacini ◽  
Lorenzo Natale
Keyword(s):  
Kalman Filter ◽  
Extended Kalman Filter ◽  
Tracking Error ◽  
Object Perception ◽  
Tactile Feedback ◽  
Tactile Sensors ◽  
Object State ◽  
Supplementary Material ◽  
Object Pose Estimation ◽  
Average Position

Tactile sensing represents a valuable source of information in robotics for perception of the state of objects and their properties. Modern soft tactile sensors allow perceiving orthogonal forces and, in some cases, relative motions along the surface of the object. Detecting and measuring this kind of lateral motion is fundamental to react to possibly uncontrolled slipping and sliding of the object being manipulated. Object slip detection and prediction have been extensively studied in the robotic community leading to solutions with good accuracy and suitable for closed-loop grip stabilization. However, algorithms for object perception, such as in-hand object pose estimation and tracking algorithms, often assume no relative motion between the object and the hand and rarely consider the problem of tracking the pose of the object subjected to slipping and sliding motions. In this work, we propose a differentiable Extended Kalman filter that can be trained to track the position and the velocity of an object under translational sliding regime from tactile observations alone. Experiments with several objects, carried out on the iCub humanoid robot platform, show that the proposed approach allows achieving an average position tracking error in the order of 0.6 cm, and that the provided estimate of the object state can be used to take control decisions using tactile feedback alone. A video of the experiments is available as Supplementary Material.

scholarly journals Combining Self-Organizing and Graph Neural Networks for Modeling Deformable Objects in Robotic Manipulation

2020 ◽  
Vol 7 ◽  
Author(s):  
Angel J. Valencia ◽  
Pierre Payeur
Keyword(s):  
Robotic Manipulation ◽  
Sensor Data ◽  
Machine Learning Techniques ◽  
Robotic Hand ◽  
Deformable Objects ◽  
Object Shape ◽  
Learning Techniques ◽  
Data Driven Approach ◽  
Graph Neural Networks ◽  
Analytical Approaches

Modeling deformable objects is an important preliminary step for performing robotic manipulation tasks with more autonomy and dexterity. Currently, generalization capabilities in unstructured environments using analytical approaches are limited, mainly due to the lack of adaptation to changes in the object shape and properties. Therefore, this paper proposes the design and implementation of a data-driven approach, which combines machine learning techniques on graphs to estimate and predict the state and transition dynamics of deformable objects with initially undefined shape and material characteristics. The learned object model is trained using RGB-D sensor data and evaluated in terms of its ability to estimate the current state of the object shape, in addition to predicting future states with the goal to plan and support the manipulation actions of a robotic hand.

scholarly journals Robotic hand illusion with tactile feedback: Unravelling the relative contribution of visuotactile and visuomotor input to the representation of body parts in space

PLoS ONE ◽  
2019 ◽  
Vol 14 (1) ◽  
pp. e0210058 ◽  
Author(s):  
The Vu Huynh ◽  
Robin Bekrater-Bodmann ◽  
Jakob Fröhner ◽  
Joachim Vogt ◽  
Philipp Beckerle
Keyword(s):  
Tactile Feedback ◽  
Robotic Hand ◽  
Body Parts ◽  
Relative Contribution

scholarly journals Dexterous Manipulation of Unknown Objects Using Virtual Contact Points

Robotics ◽  
2019 ◽  
Vol 8 (4) ◽  
pp. 86 ◽  
Author(s):  
Andrés Montaño ◽  
Raúl Suárez
Keyword(s):  
Physical Property ◽  
Robotic Hand ◽  
Tactile Sensors ◽  
Dexterous Manipulation ◽  
Everyday Objects ◽  
Simple Strategy ◽  
Contact Points ◽  
Different Shapes ◽  
Unknown Objects ◽  
Kinematic Information

The manipulation of unknown objects is a problem of special interest in robotics since it is not always possible to have exact models of the objects with which the robot interacts. This paper presents a simple strategy to manipulate unknown objects using a robotic hand equipped with tactile sensors. The hand configurations that allow the rotation of an unknown object are computed using only tactile and kinematic information, obtained during the manipulation process and reasoning about the desired and real positions of the fingertips during the manipulation. This is done taking into account that the desired positions of the fingertips are not physically reachable since they are located in the interior of the manipulated object and therefore they are virtual positions with associated virtual contact points. The proposed approach was satisfactorily validated using three fingers of an anthropomorphic robotic hand (Allegro Hand), with the original fingertips replaced by tactile sensors (WTS-FT). In the experimental validation, several everyday objects with different shapes were successfully manipulated, rotating them without the need of knowing their shape or any other physical property.

scholarly journals Grip Stabilization through Independent Finger Tactile Feedback Control

Sensors ◽  
2020 ◽  
Vol 20 (6) ◽  
pp. 1748
Author(s):  
Filipe Veiga ◽  
Benoni Edin ◽  
Jan Peters
Keyword(s):  
Grip Force ◽  
Tactile Feedback ◽  
Stabilization Method ◽  
Robotic Hands ◽  
Grip Force Control ◽  
External Perturbations ◽  
Grasp Stability ◽  
Contact Models ◽  
Biomimetic Approach ◽  
Local Slip

Grip force control during robotic in-hand manipulation is usually modeled as a monolithic task, where complex controllers consider the placement of all fingers and the contact states between each finger and the gripped object in order to compute the necessary forces to be applied by each finger. Such approaches normally rely on object and contact models and do not generalize well to novel manipulation tasks. Here, we propose a modular grip stabilization method based on a proposition that explains how humans achieve grasp stability. In this biomimetic approach, independent tactile grip stabilization controllers ensure that slip does not occur locally at the engaged robot fingers. Local slip is predicted from the tactile signals of each fingertip sensor i.e., BioTac and BioTac SP by Syntouch. We show that stable grasps emerge without any form of central communication when such independent controllers are engaged in the control of multi-digit robotic hands. The resulting grasps are resistant to external perturbations while ensuring stable grips on a wide variety of objects.

scholarly journals Bioinspired Fingertip for Anthropomorphic Robotic Hands

2014 ◽  
Vol 11 (1-2) ◽  
pp. 25-38 ◽  
Author(s):  
Marco Controzzi ◽  
Marco D'Alonzo ◽  
Carlo Peccia ◽  
Calogero Maria Oddo ◽  
Maria Chiara Carrozza ◽  
...  
Keyword(s):  
Mechanical Response ◽  
Polymeric Materials ◽  
Fe Model ◽  
Robotic Hand ◽  
Compression Tests ◽  
Robotic Hands ◽  
Experimental Conditions ◽  
Grasp Stability ◽  
Significant Step ◽  
Human Finger

Background: An artificial fingertip with mechanical features and appearance similar to the human fingertip could represent a significant step forward towards the development of the next generation artificial hands. However, so far, a fingertip showing a good trade-off among mechanical features, appearance and anthropomorphism, along with its 3D computational model, is still missing.Objective: To explore and develop an artificial fingertip demonstrating a mechanical response similar to the human fingertip, in order to improve the grasp stability of robotic hands.Methods: Taking inspiration from the multi-layered structure of the human finger, novel artificial fingertips, composed of a rigid core and covered by layers of polymeric materials with different degrees of stiffness and topped by a hard nail were developed. An accurate 3D finite element (FE) model was also developed in order to simulate and evaluate the internal mechanical behavior of the prototypes under external indentations. The mechanical response of the prototypes was assessed and compared with that of the human fingertip and the FE model results, under different experimental conditions. Finally, the artificial fingertips were integrated into an anthropomorphic robotic hand and evaluated in grip tests, in order to compare the grasp stability with respect to conventional stiff (metal) fingertips.Results: The developed prototypes demonstrated a response to compression tests similar to the human finger and the FE model showed a discrete accuracy (mean error 7%). Finally, an increased ability (by 96%) in stably holding objects during precision grips with respect to conventional stiff fingers was demonstrated.Conclusion: Multi-layered biomimetic fingertips can improve grasp stability and cosmetic appearance of anthropomorphic robot hands.

Transforming Human Hand Motion for Telemanipulation

1992 ◽  
Vol 1 (1) ◽  
pp. 63-79 ◽  
Author(s):  
Thomas H. Speeter
Keyword(s):  
Force Feedback ◽  
Robotic Manipulation ◽  
Human Hand ◽  
Robotic Hand ◽  
Robotic Hands ◽  
Automated Control ◽  
Functional Transformation ◽  
Hand Motion ◽  
Computationally Expensive ◽  
Human Hands

Manipulation by teleoperation (telemanipulation) offers an apparently straightforward and less computationally expensive route toward dextrous robotic manipulation than automated control of multifingered robotic hands. The functional transformation of human hand motions into equivalent robotic hand motions, however, presents both conceptual and analytical problems. This paper reviews and proposes algorithmic methods for transforming the actions of human hands into equivalent actions of slave multifingered robotic hands. Forward positional transformation is considered only, the design of master devices, feedforward dynamics, and force feedback are not considered although their importance for successful telemanipulation is understood. Linear, nonlinear, and functional mappings are discussed along with performance and computational considerations.

scholarly journals Development and Grasp Stability Estimation of Sensorized Soft Robotic Hand

2021 ◽  
Vol 8 ◽  
Author(s):  
P. M. Khin ◽  
Jin H. Low ◽  
Marcelo H. Ang ◽  
Chen H. Yeow
Keyword(s):  
State Estimation ◽  
Dynamic Change ◽  
Training Data ◽  
Estimation Accuracy ◽  
Robotic Hand ◽  
Stability Estimation ◽  
Pneumatic Pressure ◽  
Grasp Stability ◽  
Lstm Network ◽  
The Given

This paper introduces the development of an anthropomorphic soft robotic hand integrated with multiple flexible force sensors in the fingers. By leveraging on the integrated force sensing mechanism, grip state estimation networks have been developed. The robotic hand was tasked to hold the given object on the table for 1.5 s and lift it up within 1 s. The object manipulation experiment of grasping and lifting the given objects were conducted with various pneumatic pressure (50, 80, and 120 kPa). Learning networks were developed to estimate occurrence of object instability and slippage due to acceleration of the robot or insufficient grasp strength. Hence the grip state estimation network can potentially feedback object stability status to the pneumatic control system. This would allow the pneumatic system to use suitable pneumatic pressure to efficiently handle different objects, i.e., lower pneumatic pressure (50 kPa) for lightweight objects which do not require high grasping strength. The learning process of the soft hand is made challenging by curating a diverse selection of daily objects, some of which displays dynamic change in shape upon grasping. To address the cost of collecting extensive training datasets, we adopted one-shot learning (OSL) technique with a long short-term memory (LSTM) recurrent neural network. OSL aims to allow the networks to learn based on limited training data. It also promotes the scalability of the network to accommodate more grasping objects in the future. Three types of LSTM-based networks have been developed and their performance has been evaluated in this study. Among the three LSTM networks, triplet network achieved overall stability estimation accuracy at 89.96%, followed by LSTM network with 88.00% and Siamese LSTM network with 85.16%.

Sign in / Sign up

Export Citation Format

Share Document