Robust Haptic Rendering Using a Nonlinear Impedance Matching Approach (NIMA) for Robotic Laparoscopic Surgery

TL;DR

RAMIS haptic feedback is challenged by complex tool–tissue interactions and safety constraints. The paper introduces NIMA, a nonlinear impedance matching approach that identifies nonlinear impedance in real time to render accurate 3D forces in teleoperation. Validation shows NIMA achieves MAE $0.01$ N (SD $0.02$ N) in force rendering, a 95% improvement over linear IMA, and eliminates haptic kickback when the handle is released, enhancing safety and user comfort. These results indicate a practical path to realistic, safe haptic interfaces for robotic surgery and training.

Abstract

Background: The integration of haptic feedback into robot-assisted minimally invasive surgery (RAMIS) has long been limited by challenges in accurately rendering forces and ensuring system safety. The need for robust, high-fidelity haptic systems is critical for enhancing the precision and reliability of teleoperated surgical tools. Methods: In this study, we present a Nonlinear Impedance Matching Approach (NIMA) designed to improve force rendering by accurately modelling complex tool-tissue interactions. Based on our previously validated Impedance Matching Approach (IMA), our novel NIMA method includes nonlinear dynamics to capture and render tool-tissue forces effectively. Results: NIMA improves force feedback accuracy with a mean absolute error (MAE) of 0.01 (SD 0.02) N, achieving a 95% reduction in MAE compared to IMA. Furthermore, NIMA effectively eliminates haptic "kickback" by ensuring no force is applied by the haptic device to the user's hand when they release the handle, enhancing both patient safety and user comfort. Conclusion: NIMA's ability to account for nonlinearities in tool-tissue interactions provides an improvement in force fidelity, responsiveness, and precision across various surgical conditions. Our findings promote the advancement of haptic feedback systems for robotic surgery, offering a realistic and reliable interface for robot-assisted surgical procedures.

Figures (10)

• Figure 1: Proposed Nonlinear Impedance Matching Approach (NIMA). The leader module (left) receives motion commands $\mathbf{X}$ from the operator and renders the estimated feedback force $\mathbf{f_d}=\mathbf{M}\mathbf{X}$, while the follower module (right) executes $\mathbf{X}$, measures $(\mathbf{f},\mathbf{X})$, and identifies nonlinear impedance parameters $\mathbf{M}$ in real time. This closed-loop structure enables stable, high-fidelity force feedback in robotic laparoscopy mazidi2024nonlinear.
• Figure 2: Experimental platform used to evaluate the proposed NIMA framework. (a) The leader module consists of a dual-haptic console (Omega.7, Force Dimension) used by the surgeon to teleoperate the robotic arms through motion commands, with real-time feedback displayed on a monitor. (b) The follower module comprises two Kinova Gen3 robotic arms equipped with 6-DoF Bota force–torque sensors, custom instrument adapters, and a translucent mannequin containing a soft-tissue surrogate. The setup replicates a realistic minimally invasive surgical environment for evaluating tool–tissue interaction forces, motion tracking, and haptic feedback performance.
• Figure 3: System architecture of the proposed NIMA-enabled robot-assisted laparoscopy setup. The leader module (left) includes dual haptic devices (Omega.7) that transmit motion commands and receive force feedback estimated by the NIMA. The follower module (right) consists of two robotic arms equipped with 6-DoF force--torque sensors, an endoscope, and optical markers tracked by an NDI system for spatial registration. Force and motion data are processed through a NN for tool-tip force extraction and used by NIMA to render stable, high-fidelity haptic.
• Figure 4: Setup of Experiment 1, designed to determine the transformation between the coordinate systems of the two force sensors ($S_1$ and $S_2$). A flexible tissue representative was mounted on a Bota 6-DoF force–torque sensor ($S_2$), while the second sensor ($S_1$) was attached to the robotic arm’s end effector. The upper section of the mannequin was removed to eliminate friction forces at the RCM, allowing the measurement of pure tool–tissue interaction forces for coordinate calibration.
• Figure 5: Internal view of the mannequin used in Experiment 2, illustrating the setup for acquiring tool–tissue interaction data to train the NN model. A surgical instrument, inserted through the upper access port, interacts with a flexible tissue representative mounted on a 6-DoF Bota force–torque sensor. An endoscope provides visual feedback to monitor tool motion and contact inside the surgical workspace.