Method and Apparatus for Estimating Tremor Signals for Real-Time Application, and Robotic Surgery System Using the Same

Robotic Surgery System Estimating Tremor Signals via BMFLC-Kalman Algorithm

This technology is a control algorithm for accelerometer-based robotic surgery systems that estimates and isolates physiological tremors in real-time. By integrating Recursive Least Squares (RLS) or a Kalman filter into the BMFLC algorithm, it separates and compensates for conscious movements and unconscious tremors in a single step without the need for a pre-filter.

Existing WFLC algorithms suffer from performance degradation in real-time robotic surgery environments due to issues such as modulated frequency tracking, sensitivity to high-frequency noise, and time delays caused by the use of pre-filters.

This technology improves convergence speed and accuracy by incorporating RLS or Kalman filter operations into the BMFLC weight update stage. It minimizes latency by eliminating the need for pre-filters and ensures that only conscious movements, excluding the estimated unconscious tremor signals, are transmitted to the robot's drive unit.

‍

‍

Key Features:

• A step of measuring the operator's hand tremors using an accelerometer to obtain measurement data regarding the tremors
• A step of applying the measurement data to the BMFLC-Kalman algorithm or BMFLC-RLS algorithm to mathematically estimate the tremor signal
• A measurement unit that acquires real-time measurement data of the operator's hand tremors during surgery
• A tremor signal estimation unit that applies the measurement data to the BMFLC-Kalman or BMFLC-RLS algorithm to perform mathematical estimation

‍