CORTEXA
← Browse
arxivcs.RO2026-07-17

Data and Learning Where it Matters for Contact-Rich Manipulation

Oliver Hausdörfer, Linus Schwarz, Gabor Marko, Christian Dietz, Timo Class, Luka Hofer, Jim Yun-Jin Li, Johannes Hechtl, Ralf Römer, Angela P. Schoellig

Learned policies trained end-to-end on large datasets often remain brittle in high-precision tasks and struggle with generalization. We find that these limitations largely stem from a lack of structure and focus in data collection. Our key insight is to leverage dense data collection only for the critical segment of contact-rich tasks and to rely on traditional planning during simple free-space motion. We propose an automated data-collection scheme in combination with offline deep reinforcement learning for the critical segment of the task, eliminating reliance on a teleoperator's skill and on online policy updates. Across four challenging real-world tasks, using only 2 to 2.5 hours of autonomous data collection, we achieve an average success rate of 96%, compared to the strongest baseline at 55%. Notably, performance remains high in out-of-distribution scenarios where end-to-end approaches struggle. Our results pave the way for targeted data collection for contact-rich tasks and for high success rates in precision applications.

View free PDFSource page

Related papers

arxivcs.RO2026-06-25

VibeAct: Vibration to Actions for Contact-Rich Reactive Robot Dexterity

Yuemin Mao, Uksang Yoo, Jean Oh, Jonathan Francis, Jeffrey Ichnowski

Dexterous manipulation depends on contact events that are fast, local, and often visually occluded. Piezoelectric microphones offer a compact and high-bandwidth way to sense these interactions, but the resulting vibro-acoustic signals are difficult to simulate faithfully enough f…

View free PDFSource page
arxivcs.RO2026-07-16

Representation-Aligned Tactile Grounding for Contact-Rich Robotic Manipulation

Ruilin Chen, Jingkai Jia, Tong Yang, Xinyu Zhou, Qiao Sun, Jiangwei Zhong, et al.

Tactile-enhanced vision-language-action (VLA) policies have been introduced for contact-rich manipulation, where critical interaction states are often hidden from vision. Future tactile prediction is a promising way to use touch because it turns tactile outcomes into supervision…

View free PDFSource page
arxivcs.RO2026-07-02

VT-WAM: Visual-Tactile World Action Model for Contact-Rich Manipulation

Shuai Tian, Yupeng Zheng, Yuhang Zheng, Songen Gu, Yujie Zang, Yuxing Qin, et al.

Contact-rich manipulation requires policies to react to local deformation, pressure, slip, and friction, yet these cues are temporally sparse and often invisible in visual observations. Existing visual-tactile policies usually feed tactile observations directly into action predic…

View free PDFSource page
arxivcs.RO2026-07-03

Feeling the Unexpected: ResTacVLA for Contact-Rich Manipulation via Residual Tactile Representation

Pengwei Zhang, Bin Xie, Xinpan Meng, Xinyu Guo, Ce Hao, Fang Deng, et al.

Tactile perception is indispensable for contact-rich manipulation, yet integrating it into Vision-Language-Action (VLA) models often induces modality collapse, where high-bandwidth visual features overshadow sparse tactile cues. Inspired by Predictive Coding, a neural mechanism w…

View free PDFSource page
arxivcs.RO2026-07-20

FM-VLA: Force-based Memory for Vision-Language-Action Models in Contact-Rich Manipulation

Ruicheng Li, Qixiu Li, Ruichun Ma, Yu Deng, Lin Luo, Zhiying Du, et al.

Vision-language-action (VLA) models have achieved impressive generalization in robotic manipulation, and recent memory-augmented VLAs have relaxed the Markovian assumption by conditioning on past images or language summaries. Vision-based memory approaches address this by conditi…

View free PDFSource page
arxivcs.RO2026-07-31

TRACT: Temporally Routed Action Chunks with Chronological Phase Authority for Contact-Rich Manipulation

Jiahao Liu, Kento Kawaharazuka, Tasuku Makabe, Kei Okada

Action chunking shortens the effective decision horizon of robot imitation learning by predicting multiple future actions, while conventional phase conditioning describes the current control instant. When a predicted horizon crosses a procedural boundary, assigning the current ph…

View free PDFSource page