Skip to content
arXiv Robotics — research abstracts· Xian Nie, Yujie Zang, Yuhang Zheng, Yupeng Zheng, Songen Gu, Wendi Chen, Chuan Wen, Cewu Lu, Wenchao Ding, Junchi Yan, Shuicheng Yan·· 5 days agoEditorial score62

Contact Trajectory Prompting: In-Context Transfer of Contact-Rich Behaviors from a Single Demonstration

Contact Trajectory Prompting: In-Context Transfer of Contact-Rich Behaviors from a Single Demonstration

Summary

The paper presents Contact Trajectory Prompting (CTP), a framework that enables robots to learn contact-rich manipulation behaviors from a single demonstration. CTP encodes motion and contact information into an ordered trajectory prompt, allowing robots to perform tasks such as wiping, rotation, and insertion with high success rates. The method outperforms existing baselines by 13 percentage points in real-world tasks.

Source: arXiv Robotics — research abstracts · Read original article ↗

Loading article text…

Source:arXiv Robotics — research abstracts · arxiv.org

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).