Object-Centric Grasping Transferability: Linking Meshes to Postures

Hidalgo-Carvajal, Diego; Valle, Carlos; Naceri, Abdeldjallil; Haddadin, Sami

doi:10.1109/humanoids53995.2022.10000192

Cited by 2 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The grasping types were selected based on our grasping predictor proposed in [12]. Due to the geometric variance of unseen objects, some regions of the objects could be incorrectly labeled as non-graspable regions, or with an outlier grasping type, which was unfeasible to accomplish (c.f.…”

Section: Discussionmentioning

confidence: 99%

“…Specifically, we focused on drinking cups, bottles, and bowls. Our grasping posture predictor, proposed in [12], was used to infer grasping postures in the entire geometry of objects after shape completion. Finally, our robot hand-arm system grasped the objects in areas of interest with predefined anthropomorphic grasping sequences using our proposed hand grasping posture selection solver.…”

Section: Methodsmentioning

confidence: 99%

“…Following the insights for grasping transferability in [12], we adopted our previously proposed grasping posture predictor to select plausible grasping postures for our robot arm-hand system to approach and grasp objects. The grasping posture predictor was also parameterized by a deep neural network denoted as f posture (•).…”

Section: Grasping Posture Predictionmentioning

confidence: 99%

“…The known objects were represented by a triangular mesh with an average triangle side length of 2.5 mm. We followed the same training protocol as [12].…”

Section: E Network Training Protocolmentioning

confidence: 99%

“…• An application in robotic grasping for partial-view conditioned shape completion with shape confidence. • We leverage transferred grasping knowledge from one to multiple instances as proposed in our previous work [12]. • We further design a grasping solver for a 7-DoF robot arm in combination with under-actuated tendon-driven anthropomorphic robot hands.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Anthropomorphic Grasping With Neural Object Shape Completion

Hidalgo-Carvajal,

Chen,

Bettelani

et al. 2023

IEEE Robot. Autom. Lett.

Self Cite

View full text Add to dashboard Cite

The progressive prevalence of robots in humansuited environments has given rise to a myriad of object manipulation techniques, in which dexterity plays a paramount role. It is well-established that humans exhibit extraordinary dexterity when handling objects. Such dexterity seems to derive from a robust understanding of object properties (such as weight, size, and shape), as well as a remarkable capacity to interact with them. Hand postures commonly demonstrate the influence of specific regions on objects that need to be grasped, especially when objects are partially visible. In this work, we leverage human-like object understanding by reconstructing and completing their full geometry from partial observations, and manipulating them using a 7-DoF anthropomorphic robot hand. Our approach has significantly improved the grasping success rates of baselines with only partial reconstruction by nearly 30% and achieved over 150 successful grasps with three different object categories. This demonstrates our approach's consistent ability to predict and execute grasping postures based on the completed object shapes from various directions and positions in real-world scenarios. Our work opens up new possibilities for enhancing robotic applications that require precise grasping and manipulation skills of real-world reconstructed objects.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

Section: Grasping Posture Predictionmentioning

confidence: 99%

“…The known objects were represented by a triangular mesh with an average triangle side length of 2.5 mm. We followed the same training protocol as [12].…”

Section: E Network Training Protocolmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations