Suction helps in a pinch: Improving underwater manipulation with gentle suction flow

Stuart, Hannah; Bagheri, Matteo; Wang, Yunlong; Barnard, Heather; Sheng, Audrey L.; Jenkins, Merritt; Cutkosky, Mark R.

doi:10.1109/iros.2015.7353683

Cited by 17 publications

(11 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) Suction Contact Model: Many suction contact models acknowledge normal forces, vacuum forces, tangential friction, and torsional friction [2], [17], [23], [29] similar to a point contact with friction or soft finger model [24]. However, under this model, a single suction cup cannot resist torques about axes in the contact tangent plane, implying that any torque about such axes would cause the suction cup to drop an object (see the supplementary material for a detailed proof).…”

Section: B Wrench Space Analysismentioning

confidence: 99%

Dex-Net 3.0: Computing Robust Vacuum Suction Grasp Targets in Point Clouds Using a New Analytic Model and Deep Learning

Mahler¹,

Matl²,

Liu³

et al. 2018

2018 IEEE International Conference on Robotics and Automation (ICRA)

263

199

View full text Add to dashboard Cite

Vacuum-based end effectors are widely used in industry and are often preferred over parallel-jaw and multifinger grippers due to their ability to lift objects with a single point of contact. Suction grasp planners often target planar surfaces on point clouds near the estimated centroid of an object. In this paper, we propose a compliant suction contact model that computes the quality of the seal between the suction cup and local target surface and a measure of the ability of the suction grasp to resist an external gravity wrench. To characterize grasps, we estimate robustness to perturbations in end-effector and object pose, material properties, and external wrenches. We analyze grasps across 1,500 3D object models to generate Dex-Net 3.0, a dataset of 2.8 million point clouds, suction grasps, and grasp robustness labels. We use Dex-Net 3.0 to train a Grasp Quality Convolutional Neural Network (GQ-CNN) to classify robust suction targets in point clouds containing a single object. We evaluate the resulting system in 350 physical trials on an ABB YuMi fitted with a pneumatic suction gripper. When evaluated on novel objects that we categorize as Basic (prismatic or cylindrical), Typical (more complex geometry), and Adversarial (with few available suction-grasp points) Dex-Net 3.0 achieves success rates of 98%, 82%, and 58% respectively, improving to 81% in the latter case when the training set includes only adversarial objects. Code, datasets, and supplemental material can be found at http://berkeleyautomation.github.io/dex-net.

show abstract

Section: B Wrench Space Analysismentioning

confidence: 99%

Dex-Net 3.0: Computing Robust Vacuum Suction Grasp Targets in Point Clouds Using a New Analytic Model and Deep Learning

Mahler¹,

Matl²,

Liu³

et al. 2018

2018 IEEE International Conference on Robotics and Automation (ICRA)

263

199

View full text Add to dashboard Cite

show abstract

“…Limpet-inspired suction cups have been applied for artifact gathering in the deep-sea, such as at shipwreck sites (Søreide, 2011 ). Suction flow was incorporated onto a multi-finger underwater hand (Stuart et al, 2019 ), and the monitoring of suction flows has been introduced as a potential way to perform tactile sensing under water (Stuart et al, 2015 ). Even without suction pumps, water drag effects on objects has been shown to improve the capture of floating objects, as compared to grasping in the vacuum of space (Stuart et al, 2019 ).…”

Section: Hands In the Fieldmentioning

confidence: 99%

Hands in the Real World

2020

Self Cite

View full text Add to dashboard Cite

Robots face a rapidly expanding range of potential applications beyond controlled environments, from remote exploration and search-and-rescue to household assistance and agriculture. The focus of physical interaction is typically delegated to end-effectors-fixtures, grippers or hands-as these machines perform manual tasks. Yet, effective deployment of versatile robot hands in the real world is still limited to few examples, despite decades of dedicated research. In this paper we review hands that found application in the field, aiming to discuss open challenges with more articulated designs, discussing novel trends and perspectives. We hope to encourage swift development of capable robotic hands for long-term use in varied real world settings. The first part of the paper centers around progress in artificial hand design, identifying key functions for a variety of environments. The final part focuses on the overall trends in hand mechanics, sensors and control, and how performance and resiliency are qualified for real world deployment.

show abstract

“…The tendon-driven Red Sea Exploratorium hand uses one motor per finger to achieve multiple grasp types for use with tools and biological specimens (Stuart et al, 2014(Stuart et al, , 2015. The elastic finger joints have low bending stiffness, and use substantial preloads to dictate finger curling behavior without diminishing grasp strength.…”

Section: Compliant Hands For Marine Applicationsmentioning

confidence: 99%

The Ocean One hands: An adaptive design for robust marine manipulation

Stuart

Wang

Khatib

et al. 2017

The International Journal of Robotics Research

Self Cite

135

View full text Add to dashboard Cite

Underactuated, compliant, tendon-driven robotic hands are suited for deep-sea exploration. The robust Ocean One hand design utilizes elastic finger joints and a spring transmission to achieve a variety of pinch and wrap grasps. Compliance in the fingers and transmission determines the degree of load-sharing among contacts and the hands' ability to secure irregularly shaped objects. However, it can also decrease external grasp stiffness and acquisition reliability. SimGrasp, a flexible dynamic hand simulator, enables parametric studies of the hand for acquisition and pull-out tests with varying transmission spring rates. In the present application, we take advantage of achieving different stiffnesses by reversing the direction of tendon windup using a torsional spring-loaded winch. With this provision, the hand can be relatively soft for handling delicate objects and stiff for tasks requiring strength. Two hands were field-tested as part of the Ocean One humanoid platform, which acquired a vase from the La Lune shipwreck site at a 91 m depth in the Mediterranean Sea.

show abstract

Suction helps in a pinch: Improving underwater manipulation with gentle suction flow

Cited by 17 publications

References 17 publications

Dex-Net 3.0: Computing Robust Vacuum Suction Grasp Targets in Point Clouds Using a New Analytic Model and Deep Learning

Dex-Net 3.0: Computing Robust Vacuum Suction Grasp Targets in Point Clouds Using a New Analytic Model and Deep Learning

Hands in the Real World

The Ocean One hands: An adaptive design for robust marine manipulation

Contact Info

Product

Resources

About