Trajectory Divergence Horizon Decision for Reliable Dual-Arm Surgical Subtask Manipulation

2026-08-10

AI Summary

Current vision-language-action (VLA) models in surgical robotics rely on fixed-length, open-loop action sequences, which struggle to adapt to dynamic environments, often accumulating errors and introducing safety risks. This work proposes a Trajectory Divergence-based Horizon Decision (TDHD) mechanism that, for the first time, introduces a divergence metric derived from dual-stream trajectory matching under minimal noise perturbations. Coupled with a dual-threshold rule, TDHD dynamically truncates execution to trigger replanning, enabling real-time reliability assessment and adaptive control during inference. Evaluated on a custom-built dual-arm da Vinci–style surgical platform featuring synchronized multi-view perception and language instructions, the method significantly improves task success rates—increasing needle handling from 55% to 60% and tissue manipulation from 55% to 80%, with particularly notable gains in late-stage task execution—thereby enhancing the cross-task reusability and safety of VLA models.

Abstract

Surgical robotic systems are increasingly being adopted as clinical workload rises, motivating autonomous solutions for repetitive manipulation subtasks. Learning-based controllers improve generalization compared with rule-based and analytic approaches, but most are trained for individual tasks and remain difficult to reuse across procedures. Vision-Language-Action (VLA) models provide a unified framework that integrates visual perception, language grounding, and action generation, offering a promising path toward more composable surgical autonomy. However, existing VLA policies rely on fixed-length open-loop action sequences, where changing scene conditions can lead to accumulated errors and potential risks in surgical manipulation. To mitigate this issue, we formulate surgical VLA deployment as an adaptive execution-horizon decision problem and propose Trajectory Divergence Horizon Decision (TDHD), a test-time mechanism that estimates step-wise action reliability by measuring the divergence between two flow-matching-generated trajectories under small noise perturbations and truncates execution using a dual-threshold rule to trigger timely replanning. We further establish a real-world da Vinci-like dual-arm benchmark with synchronized multi-view perception and language instructions, and collect 600 teleoperated demonstrations across needle (reach, pick, regrasp) and tissue (reach, lift, resection) manipulation suites. On real hardware with 20 trials per task setting, TDHD consistently improves performance over the latest VLA baselines: success increases from 55\% to 60\% for needle manipulation and from 55\% to 80\% for tissue manipulation, with the largest gains observed in the final manipulation stages. These results highlight the importance of adaptive execution control for reliable deployment of VLA models in surgical robotic manipulation.

Problem

• surgical robotics
• Vision-Language-Action models
• trajectory divergence
• adaptive execution
• reliable manipulation

Innovation

• Trajectory Divergence Horizon Decision
• Vision-Language-Action models
• adaptive execution horizon
• surgical robotics
• flow-matching trajectories