Grasp-Taxonomy Conditioning as a Mode Switch for Dexterous Grasping of Climbing Holds
Overview
This paper introduces BetaGrasp, a 3D diffusion policy that uses a four-class grasp label (crimp, sloper, pinch, jug) to condition a DP3-style point-cloud policy on a Franka arm with a 16-DoF LEAP hand. The authors also present a climbing-hold benchmark using a spring testbed with ratchet slip-force readings. Analysis shows the point-cloud encoder alone fails to distinguish hold types, leading to incorrect jug grasps in simulation without the label; BetaGrasp achieves 94% correct grasp type versus 25% for the baseline. On 80 hardware trials, BetaGrasp increases average slip force from 9.6 N to 14.3 N and reduces fully slipped grasps from 44% to 10% of trials.
Generated from attributed reports · Updated 7 hours ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · average slip force: 9.6 other · Basis not reported · Differing source assertions
“cy performs. On 80 paired hardware trials, BetaGrasp raises average slip force from 9.6 N to 14.3 N, and grasps slipping fully off the hold drop from 44% of trials to 10%.”
Exact source · revision 1Source owner not reported
Reported quantity · average slip force: 14.3 other · Basis not reported · Differing source assertions
“cy performs. On 80 paired hardware trials, BetaGrasp raises average slip force from 9.6 N to 14.3 N, and grasps slipping fully off the hold drop from 44% of trials to 10%.”
Exact source · revision 1Source owner not reported
Reported quantity · hardware trials: 80 other · reported trials
“cy performs. On 80 paired hardware trials, BetaGrasp raises average slip force from 9.6 N to 14.3 N, and grasps slipping fully off the hold drop from 44% of trials to 10%.”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalGrasp-Taxonomy Conditioning as a Mode Switch for Dexterous Grasping of Climbing Holds
This paper introduces BetaGrasp, a 3D diffusion policy that uses a four-class grasp label (crimp, sloper, pinch, jug) to condition a DP3-style point-cloud policy on a Franka arm with a 16-DoF LEAP hand. The authors also present a climbing-hold benchmark using a spring testbed with ratchet slip-force readings. Analysis shows the point-cloud encoder alone fails to distinguish hold types, leading to incorrect jug grasps in simulation without the label; BetaGrasp achieves 94% correct grasp type versus 25% for the baseline. On 80 hardware trials, BetaGrasp increases average slip force from 9.6 N to 14.3 N and reduces fully slipped grasps from 44% to 10% of trials.
Event coverage history
There is not enough continuous observation data to show a trend.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).