An Interactive Physically-based Model for Active Suction Phenomenon Simulation

Betnardin, Antonin; Duriez, Christian; Marchal, Maud

doi:10.1109/iros40897.2019.8967526

Cited by 13 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…High complexity may lead to a long computation cost of annotation. Parameter identification by real world experiment (Bernardin et al, 2019 ) might also be necessary to ensure the validity of the contact model.…”

Section: Related Workmentioning

confidence: 99%

Learning Suction Graspability Considering Grasp Quality and Robot Reachability for Bin-Picking

Jiang

Oaki²,

Ishihara³

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

Deep learning has been widely used for inferring robust grasps. Although human-labeled RGB-D datasets were initially used to learn grasp configurations, preparation of this kind of large dataset is expensive. To address this problem, images were generated by a physical simulator, and a physically inspired model (e.g., a contact model between a suction vacuum cup and object) was used as a grasp quality evaluation metric to annotate the synthesized images. However, this kind of contact model is complicated and requires parameter identification by experiments to ensure real world performance. In addition, previous studies have not considered manipulator reachability such as when a grasp configuration with high grasp quality is unable to reach the target due to collisions or the physical limitations of the robot. In this study, we propose an intuitive geometric analytic-based grasp quality evaluation metric. We further incorporate a reachability evaluation metric. We annotate the pixel-wise grasp quality and reachability by the proposed evaluation metric on synthesized images in a simulator to train an auto-encoder–decoder called suction graspability U-Net++ (SG-U-Net++). Experiment results show that our intuitive grasp quality evaluation metric is competitive with a physically-inspired metric. Learning the reachability helps to reduce motion planning computation time by removing obviously unreachable candidates. The system achieves an overall picking speed of 560 PPH (pieces per hour).

show abstract

Section: Related Workmentioning

confidence: 99%

Learning Suction Graspability Considering Grasp Quality and Robot Reachability for Bin-Picking

Jiang

Oaki²,

Ishihara³

et al. 2022

Front. Neurorobot.

View full text Add to dashboard Cite

show abstract

“…Mahler et al [2018] also propose a model to quantify the ability of a passive suction cup to stay attached to a target surface when lifting the object. Bernardin et al [2019] propose a method to simulate active suction cups for robotics applications. In our work, we use similar geometric models, but our approach for the detection and monitoring of cavities is much more robust because we compute geometric intersections rather than the classification of vertices and flood-fill of surface regions.…”

Section: Related Workmentioning

confidence: 99%

Constraint-based Simulation of Passive Suction Cups

et al. 2022

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a physics-based model of suction phenomenon to achieve simulation of deformable objects like suction cups. Our model uses a constraint-based formulation to simulate the variations of pressure inside suction cups. The respective internal pressures are represented as pressure constraints which are coupled with anti-interpenetration and friction constraints. Furthermore, our method is able to detect multiple air cavities using information from collision detection. We solve the pressure constraints based on the ideal gas law while considering several cavity states. We test our model with a number of scenarios reflecting a variety of uses, for instance, a spring loaded jumping toy, a manipulator performing a pick and place task, and an octopus tentacle grasping a soda can. We also evaluate the ability of our model to reproduce the physics of suction cups of varying shapes, lifting objects of different masses, and sliding on a slippery surface. The results show promise for various applications such as the simulation in soft robotics and computer animation.

show abstract

“…High complexity may lead to long computation cost of annotation. Parameter identification by real world experiment [Bernardin et al, 2019] might also be necessary to ensure the validity of the contact model.…”

Section: Pixel-wise Graspability Learningmentioning

confidence: 99%

Learning suction graspability considering grasp quality and robot reachability for bin-picking

Jiang¹,

Oaki²,

Ishihara³

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep learning has been widely used for inferring robust grasps. Although human-labeled RGB-D datasets were initially used to learn grasp configurations, preparation of this kind of large dataset is expensive. To address this problem, images were generated by a physical simulator, and a physically inspired model (e.g., a contact model between a suction vacuum cup and object) was used as a grasp quality evaluation metric to annotate the synthesized images. However, this kind of contact model is complicated and requires parameter identification by experiments to ensure real world performance. In addition, previous studies have not considered manipulator reachability such as when a grasp configuration with high grasp quality is unable to reach the target due to collisions or the physical limitations of the robot. In this study, we propose an intuitive geometric analytic-based grasp quality evaluation metric. We further incorporate a reachability evaluation metric. We annotate the pixel-wise grasp quality and reachability by the proposed evaluation metric on synthesized images in a simulator to train an auto-encoder-decoder called suction graspability U-Net++ (SG-U-Net++). Experiment results show that our intuitive grasp quality evaluation metric is competitive with a physically-inspired metric. Learning the reachability helps to reduce motion planning computation time by removing obviously unreachable candidates. The system achieves an overall picking speed of 560 PPH (pieces per hour).

show abstract

An Interactive Physically-based Model for Active Suction Phenomenon Simulation

Cited by 13 publications

References 13 publications

Learning Suction Graspability Considering Grasp Quality and Robot Reachability for Bin-Picking

Learning Suction Graspability Considering Grasp Quality and Robot Reachability for Bin-Picking

Constraint-based Simulation of Passive Suction Cups

Learning suction graspability considering grasp quality and robot reachability for bin-picking

Contact Info

Product

Resources

About