ForeTac: Predictive Contact Guidance for Generative Robot Policies
ForeTac: Predictive Contact Guidance for Generative Robot Policies
ForeTac is a predictive contact-guidance framework that improves robot task success rates by anticipating and evaluating contact outcomes through tactile feedback. It uses a represent-predict-evaluate-guide loop, with a contact-quality objective learned from physical interactions. Evaluated on four real-robot tasks, ForeTac increases average task success from 36.3% to 71.3% and contact success rate from 20.0% to 66.3% compared to vision-only Diffusion Policy.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported success rate
36.3%
Reported success rate
20%
Reported success rate
71.3%
Reported success rate
66.3%
- paper: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
ide action generation while the pretrained model parameters remain frozen. We evaluate ForeTac on four real-robot tasks (planar and curved-surface wiping, card swiping, and fragile-object grasping) with diffusion and flow-matching policies. Across these tasks, ForeTac raises the average task success rate from 36.3% to 71.3% and the average contact success rate (CSR) from 20.0% to 66.3% compared with vision-only Diffu
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org