Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces OmniTacTune, a policy-agnostic reinforcement learning (RL) framework that enhances visual policies for contact-rich manipulation by integrating tactile feedback through residual adaptation. The method employs a two-stage design, first leveraging autonomous rollouts to bootstrap tactile-aware learning and subsequently refining a lightweight tactile residual policy via online interaction. Experimental results reveal that OmniTacTune significantly boosts the success rates of visual base policies across various tasks, achieving improvements from 5-40% to 85-100% in less than two hours of training.
Tactile feedback can elevate visual policies in contact-rich tasks from mediocre to near-perfect performance in under 80 minutes.
Visual policies learned from human videos, teleoperation, and robot demonstrations offer scalable motion priors, but often fail in contact-rich manipulation, where success significantly depends on local force and contact geometry. Tactile sensing provides these complementary signals, yet tactile data remain costly to collect and hard to generalize across sensors, robots, and tasks. We introduce OmniTacTune, a policy-agnostic real-world RL pipeline that adapts tactile feedback to pretrained visual policies through residual correction. OmniTacTune uses a two-stage design: it first bootstraps tactile-aware learning from autonomous base-policy rollouts, then learns a lightweight tactile residual policy through online interaction. Extensive experiments show that OmniTacTune generalizes across diverse contact-rich tasks, visual base policies, and tactile representations. Across four real-world contact-rich tasks, it improves visual base policies from 5-40% success to 85-100% within 40-80 minutes, demonstrating an efficient path for adapting tactile feedback to scalable visual robot policies. Project page: https://colinyu1.github.io/omnitactune-site/