CADeT: Causal-Aware Deformation Transmission for Indirect Robotic Manipulation of Soft Tissue

2026-09-29

AI Summary

This study addresses the challenge of ambiguous and indistinguishable deformation transmission modes during indirect manipulation of deep soft tissues in minimally invasive surgery. To this end, we propose a causality-aware control framework that integrates structural causal models with active perception strategies to infer latent transmission modes online, thereby resolving observational ambiguity and estimating the adhesion Jacobian. This inference is subsequently coupled with belief-based model predictive control to achieve precise indirect manipulation. Simulations and experiments on a da Vinci surgical robot platform demonstrate that the proposed method significantly improves transmission mode recognition accuracy and accelerates shape error convergence. Overall, this work provides an innovative solution for the complex indirect manipulation of soft tissues.

Abstract

Indirect manipulation of deep-seated deformable anatomy inaccessible to the robot is challenging in robot-assisted minimally invasive surgery (RAMIS) because intervening tissues spatially filter deformation transmission. Passive observations can be ambiguous because the Decoupled and Blocked modes may produce similar motion responses. We propose CADeT, a causal-aware deformation transmission framework that integrates structural causal model (SCM) with active sensing to infer a latent transmission mode and estimate a state-dependent adhesion Jacobian online. During normal manipulation, control actions update the mode belief; when ambiguity persists, an additional probing action is selected to improve mode distinguishability. The mode belief and learned Jacobian are incorporated into a belief-aware model predictive controller for indirect target-shape control. Validation in simulation and on the da Vinci research kit (dVRK), using phantom and ex vivo porcine tissues, shows higher mode-identification accuracy and faster shape-error convergence than the evaluated model-free and model-based baselines. These results show that active sensing improves mode identification and indirect deformation control under the evaluated conditions.

Problem

• indirect manipulation
• soft tissue
• deformation transmission
• robot-assisted minimally invasive surgery
• mode ambiguity

Innovation

• Causal-Aware Deformation Transmission
• Structural Causal Model
• Active Sensing
• Belief-Aware Model Predictive Control
• Indirect Robotic Manipulation