Visuo-tactile Diffusion Policy for Robotic Manipulation with Auxiliary Tactile Dynamics Prediction
Overview
This research introduces a visuotactile diffusion policy that combines four-step histories from two camera views and two fingertip tactile images to condition action-sequence generation for long-horizon robotic assembly. The model uses a recurrent temporal representation integrating recent observations from both modalities and an auxiliary objective that predicts fingertip-image changes two steps ahead during training to capture how contact evolves under action. Across three independent training seeds in MuJoCo assembly experiments, the full model achieves an average success rate of 59.0% ± 2.6%, improving on the visual-only baseline (42.7% ± 1.5%) by 16.3 percentage points. The auxiliary objective adds no inference-time computation.
Generated from attributed reports · Updated 2 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 · standard deviation (full model): 2.6 other · Basis not reported
“-time computation. Across three independent training seeds in MuJoCo assembly experiments, the full model achieves an average success rate of $59.0\% \pm 2.6\%$, improving on the visual-only baseline ($42.7\% \pm 1.5\%$) by 16.3 percentage points.”
Exact source · revision 1Source owner not reported
Reported quantity · standard deviation (baseline): 1.5 other · Basis not reported
“-time computation. Across three independent training seeds in MuJoCo assembly experiments, the full model achieves an average success rate of $59.0\% \pm 2.6\%$, improving on the visual-only baseline ($42.7\% \pm 1.5\%$) by 16.3 percentage points.”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsSignalVisuo-tactile Diffusion Policy for Robotic Manipulation with Auxiliary Tactile Dynamics Prediction
This research introduces a visuotactile diffusion policy that combines four-step histories from two camera views and two fingertip tactile images to condition action-sequence generation for long-horizon robotic assembly. The model uses a recurrent temporal representation integrating recent observations from both modalities and an auxiliary objective that predicts fingertip-image changes two steps ahead during training to capture how contact evolves under action. Across three independent training seeds in MuJoCo assembly experiments, the full model achieves an average success rate of 59.0% ± 2.6%, improving on the visual-only baseline (42.7% ± 1.5%) by 16.3 percentage points. The auxiliary objective adds no inference-time computation.
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).