SIMS: Surgeon-Intention-driven Motion Scaling for Efficient and Precise Teleoperation

TL;DR

This work tackles the trade‑off between precision and efficiency in telesurgery caused by fixed motion scaling factors ($MSF$). It proposes Surgeon‑Intention driven Motion Scaling (SIMS), which infers intent into one of three motion classes ($ ext{fine}$, $ ext{neutral}$, $ ext{coarse}$) using three kinematic features ($f_{1}$, $f_{2}$, $f_{3}$) fed to fuzzy C‑means clustering, with per‑arm confidence‑based MSF updates and OR fusion across arms. In a n=10 user study on the dVRK performing FLS tasks, SIMS reduces collisions by over 80% and lowers perceived workload while maintaining comparable efficiency to larger fixed scaling, demonstrating a practical, low‑latency path to safer, more efficient telesurgical control. The approach is lightweight, relies solely on kinematics, and shows promise for integration with higher‑level context recognition to further tailor scaling to task phases.

Abstract

Telerobotic surgery often relies on a fixed motion scaling factor (MSF) to map the surgeon's hand motions to robotic instruments, but this introduces a trade-off between precision and efficiency: small MSF enables delicate manipulation but slows large movements, while large MSF accelerates transfer at the cost of accuracy. We propose a Surgeon-Intention driven Motion Scaling (SIMS) system, which dynamically adjusts MSF in real time based solely on kinematic cues. SIMS extracts linear speed, tool motion alignment, and dual-arm coordination features to classify motion intent via fuzzy C-means clustering and applies confidence-based updates independently for both arms. In a user study (n=10, three surgical training tasks) conducted on the da Vinci Research Kit, SIMS significantly reduced collisions (mean reduction of 83%), lowered mental and physical workload, and maintained task completion efficiency compared to fixed MSF. These findings demonstrate that SIMS is a practical and lightweight approach for safer, more efficient, and user-adaptive telesurgical control.

Figures (5)

• Figure 1: Overview of SIMS. Conventional teleoperation relies on a fixed MSF, where a small MSF ensures precision but slows down large movements, and a large MSF allows fast movements but sacrifices accuracy (top). The proposed SIMS dynamically adjusts the MSF based on the surgeon’s intent: when coarse motion is detected, the MSF increases to enable efficient movement, and when fine motion is required, it decreases to ensure precision (bottom).
• Figure 2: System Diagram of SIMS. SIMS infers the surgeon’s intended motion scale (coarse, neutral, fine) from hand-controller inputs and dynamically adjusts the MSF. A. Feature Extraction: Real-time pose commands ($u_t$) generate trajectories, from which linear speed, tool motion alignment, and dual-arm coordination are extracted. B. Intended Scaling Estimation: Pre-trained FCM models output membership values for each motion class, which are fused to estimate the intended scale at 60 Hz during teleoperation. C. User-Adaptive Update: The inferred scale determines the ramp function’s sign and magnitude for MSF updates: fine decreases MSF, coarse increases MSF, and neutral maintains it. Features are buffered, and when sequence length and variance criteria are met, FCM models are updated with recent data to reflect user adaptation while preserving diversity.
• Figure 3: Evaluation Surgical Tasks for SIMS and Fixed MSF Comparison. Experimental tasks: (a) peg transfer, (b) phantom tissue setup, (c) surgical stitching, and (d) surgical knot tying.
• Figure 4: NASA-TLX workload survey The results across six subscales (mental, physical, temporal demand, performance, effort, frustration). SIMS shows consistently lower perceived workload compared to fixed MSF settings. * indicates $p < 0.05$, ** indicates $p < 0.01$, and ns denotes not significant.
• Figure 5: End-effector Trajectory Representative end-effector trajectories during the peg transfer task under three configurations. Fixed small scale (green, top) required frequent clutching, leading to fragmented and repetitive motion. Fixed large scale (red, bottom) caused over-amplification and jittery movements, increasing collision risk. SIMS (blue) maintained smooth, continuous trajectories with reduced clutching and improved control stability.