News

Contact Conditional Vision-based Force Estimation to be Published in JMRR

Our paper on contact-conditional vision-based force estimation is accepted for publication in the Journal of Medical Robotics Research.

Our paper [1] on contact-conditional visual force estimation using local stiffness model has been accepted for publication in the Journal of Medical Robotics Research. This is the first student-led journal paper from the lab! Congratulations to lead author Shuyuan Yang and the rest of the contributors!

Our previous work has shown that vision-based force estimation neural networks have difficulty generalizing to new scenes. Therefore we asked the question: How can we create vision-based force estimators that can quickly generalize to new scenes especially when training data is difficult to acquire?

In this work, we train neural networks without the need for large amounts of robot sensor data. Instead, we leverage the convenience of video recordings, the scalability of crowd-sourcing, and geometrically-aware neural representations to enhance data-efficiency when fine-tuning to generalize to new surgical scenes.

2024

Journal Articles

  1. S. Yang, M. H. Le, K. R. Golobish, J. C. Beaver, and Z. Chua, “Vision-Based Force Estimation for Minimally Invasive Telesurgery Through Contact Detection and Local Stiffness Models,” Journal of Medical Robotics Research, p. 2440008, 2024.

Other News