VOMMI: Collecting and Leveraging Portable Demonstrations for Mobile Manipulation
Overview
This research presents VOMMI, a framework for collecting portable demonstrations to address data scarcity in embodied intelligence. It connects RGB observations to vision-language-action (VLA) post-training through offline trajectory reconstruction and online visual-motion conditioning. The system synchronizes body and hand views without kinematic calibration, using R2-VO to refine trajectories with sparse geometric anchors. Experiments show that policies post-trained on portable demonstrations achieve 18.2% lower base-velocity error than those trained with robot-collected demonstrations, while maintaining comparable end-effector accuracy. Offline reconstruction reduces absolute trajectory errors by 24.6% on average, and the complete system improves mean success rate by 8.3 percentage points over OpenPI 0.5 across three real-robot tasks.
Generated from attributed reports · Updated 2 days 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 · trajectories: 500 other · Basis not reported
“y into the base branch. Experiments use a 500-trajectory portable for each task, with 75 trajectories held out for RGB-VO evaluation, and 200 robot demonstrations as references. Our policy, post-trained only on portable demonstrations, achieves 18.2% lower base-velocity error than a policy trained with robot-collected demonstrations, while maintaining comparable end-effector translation accuracy. Offline reconstructi”
Exact source · revision 1Source owner not reported
Reported quantity · Robot trajectories: 75 trajectories · trajectory count
“y into the base branch. Experiments use a 500-trajectory portable for each task, with 75 trajectories held out for RGB-VO evaluation, and 200 robot demonstrations as references. Our policy, post-trained only on portable demonstrations, achieves 18.2% lower base-velocity error than a policy trained with robot-collected demonstrations, while maintaining comparable end-effector translation accuracy. Offline reconstructi”
Exact source · revision 1Source owner not reported
Reported quantity · Robot count: 200 robots · Basis not reported
“y into the base branch. Experiments use a 500-trajectory portable for each task, with 75 trajectories held out for RGB-VO evaluation, and 200 robot demonstrations as references. Our policy, post-trained only on portable demonstrations, achieves 18.2% lower base-velocity error than a policy trained with robot-collected demonstrations, while maintaining comparable end-effector translation accuracy. Offline reconstructi”
Exact source · revision 1Source owner not reported
Reported quantity · percent: 18.2 other · Basis not reported · Differing source assertions
“y into the base branch. Experiments use a 500-trajectory portable for each task, with 75 trajectories held out for RGB-VO evaluation, and 200 robot demonstrations as references. Our policy, post-trained only on portable demonstrations, achieves 18.2% lower base-velocity error than a policy trained with robot-collected demonstrations, while maintaining comparable end-effector translation accuracy. Offline reconstructi”
Exact source · revision 1Source owner not reported
Reported quantity · percent: 24.6 other · Basis not reported · Differing source assertions
“on reduces absolute trajectory errors for the body and hand streams by 24.6% on average relative to the best evaluated baseline for each stream. The complete system improves the mean success rate by 8.3 percentage points over OpenPI 0.5 across three real-robot tasks.”
Exact source · revision 1Source owner not reported
Reported quantity · percent: 8.3 other · Basis not reported · Differing source assertions
“on reduces absolute trajectory errors for the body and hand streams by 24.6% on average relative to the best evaluated baseline for each stream. The complete system improves the mean success rate by 8.3 percentage points over OpenPI 0.5 across three real-robot tasks.”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalVOMMI: Collecting and Leveraging Portable Demonstrations for Mobile Manipulation
This research presents VOMMI, a framework for collecting portable demonstrations to address data scarcity in embodied intelligence. It connects RGB observations to vision-language-action (VLA) post-training through offline trajectory reconstruction and online visual-motion conditioning. The system synchronizes body and hand views without kinematic calibration, using R2-VO to refine trajectories with sparse geometric anchors. Experiments show that policies post-trained on portable demonstrations achieve 18.2% lower base-velocity error than those trained with robot-collected demonstrations, while maintaining comparable end-effector accuracy. Offline reconstruction reduces absolute trajectory errors by 24.6% on average, and the complete system improves mean success rate by 8.3 percentage points over OpenPI 0.5 across three real-robot tasks.
Event coverage history
Current attention 3·Peak within the comparable range 4（2026-11-10 14:00 UTC）·Change within the comparable range over 24 hours –
The trend compares participants observed continuously; its scope may be narrower than current attention totals. Hover or click to explore hourly observations, or use the left and right arrow keys.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).