Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Predictive Approach for Enhancing Accuracy in Remote Robotic Surgery Using Informer Model

Muhammad Hanif Lashari, Shakil Ahmed, Wafa Batayneh, Ashfaq Khokhar

TL;DR

This work tackles real-time, high-accuracy 3D position estimation for the patient-side robotic manipulator in remote surgery over the Tactile Internet, where network delays and losses threaten control fidelity. It introduces a predictive framework based on the Transformer-derived Informer model, augmented with a Four-State Hidden Markov Model to simulate realistic packet loss and a differentiable optimization layer to enforce energy, smoothness, and robustness constraints. The approach leverages ProbSparse attention, attention distilling, a fast generative decoder, and network-aware inputs to achieve stable, low-latency predictions with complexity $O(L \log L)$, and demonstrates over 90% position accuracy on the JIGSAWS knot-tying dataset, outperforming LSTM, RNN, and TCN baselines. These results indicate strong potential for TI-enabled remote robotic surgery, enabling reliable haptic-enabled control in the presence of bursty network disturbances.

Abstract

Precise and real-time estimation of the robotic arm's position on the patient's side is essential for the success of remote robotic surgery in Tactile Internet (TI) environments. This paper presents a prediction model based on the Transformer-based Informer framework for accurate and efficient position estimation. Additionally, it combines a Four-State Hidden Markov Model (4-State HMM) to simulate realistic packet loss scenarios. The proposed approach addresses challenges such as network delays, jitter, and packet loss to ensure reliable and precise operation in remote surgical applications. The method integrates the optimization problem into the Informer model by embedding constraints such as energy efficiency, smoothness, and robustness into its training process using a differentiable optimization layer. The Informer framework uses features such as ProbSparse attention, attention distilling, and a generative-style decoder to focus on position-critical features while maintaining a low computational complexity of O(L log L). The method is evaluated using the JIGSAWS dataset, achieving a prediction accuracy of over 90 percent under various network scenarios. A comparison with models such as TCN, RNN, and LSTM demonstrates the Informer framework's superior performance in handling position prediction and meeting real-time requirements, making it suitable for Tactile Internet-enabled robotic surgery.

A Predictive Approach for Enhancing Accuracy in Remote Robotic Surgery Using Informer Model

TL;DR

This work tackles real-time, high-accuracy 3D position estimation for the patient-side robotic manipulator in remote surgery over the Tactile Internet, where network delays and losses threaten control fidelity. It introduces a predictive framework based on the Transformer-derived Informer model, augmented with a Four-State Hidden Markov Model to simulate realistic packet loss and a differentiable optimization layer to enforce energy, smoothness, and robustness constraints. The approach leverages ProbSparse attention, attention distilling, a fast generative decoder, and network-aware inputs to achieve stable, low-latency predictions with complexity , and demonstrates over 90% position accuracy on the JIGSAWS knot-tying dataset, outperforming LSTM, RNN, and TCN baselines. These results indicate strong potential for TI-enabled remote robotic surgery, enabling reliable haptic-enabled control in the presence of bursty network disturbances.

Abstract

Precise and real-time estimation of the robotic arm's position on the patient's side is essential for the success of remote robotic surgery in Tactile Internet (TI) environments. This paper presents a prediction model based on the Transformer-based Informer framework for accurate and efficient position estimation. Additionally, it combines a Four-State Hidden Markov Model (4-State HMM) to simulate realistic packet loss scenarios. The proposed approach addresses challenges such as network delays, jitter, and packet loss to ensure reliable and precise operation in remote surgical applications. The method integrates the optimization problem into the Informer model by embedding constraints such as energy efficiency, smoothness, and robustness into its training process using a differentiable optimization layer. The Informer framework uses features such as ProbSparse attention, attention distilling, and a generative-style decoder to focus on position-critical features while maintaining a low computational complexity of O(L log L). The method is evaluated using the JIGSAWS dataset, achieving a prediction accuracy of over 90 percent under various network scenarios. A comparison with models such as TCN, RNN, and LSTM demonstrates the Informer framework's superior performance in handling position prediction and meeting real-time requirements, making it suitable for Tactile Internet-enabled robotic surgery.
Paper Structure (30 sections, 17 equations, 4 figures, 2 tables)

This paper contains 30 sections, 17 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Problem Statement
  4. System Model
  5. System Overview
  6. Network Errors and Challenges
  7. Optimization Problem
  8. Proposed Model
  9. Modeling Packet Loss
  10. Informer Model-based Predictive Approch
  11. Description of the Informer Model
  12. Optimized Attention Mechanism
  13. Identifying Relevant Queries
  14. ProbSparse Attention Mechanism
  15. Streamlined Self-attention Distilling
  16. ...and 15 more sections

Figures (4)

  • Figure 1: Remote Robotic Surgery Framework Utilizing TI and Informer Model for Enhanced PSM Precision
  • Figure 2: Informer Model Encoder-Decoder Framework with ProbSparse Attention Mechanism
  • Figure 3: The top plot (Part 1) shows the simulated packet loss pattern across 1000 time steps. The following plots (Parts 2-4) display the original and corrupted tool tip position along all three axes, with gray-shaded regions highlighting periods of packet loss.
  • Figure 4: Prediction performance of the Informer model under packet loss for tool tip position in X, Y, and Z axes. Solid and dashed lines represent actual and predicted positions, respectively. The model achieves accuracies of 96.68%, 95.96%, and 90.37% for the X, Y, and Z axes, demonstrating robustness against network-induced packet loss.