SCADE: Simultaneous Sensor Calibration and Deformation Estimation of FBG-Equipped Unmodeled Continuum Manipulators

Area of Science:

• Robotics
• Medical Devices
• Sensor Technology

Background:

• Continuum manipulators (CMs) are increasingly used in surgery, requiring precise shape and position estimation.
• Existing methods struggle with inaccurate embedded sensors and intermittent external data fusion.
• Accurate CM modeling is critical for safe and effective robotic interventions.

Purpose of the Study:

• To introduce a novel stochastic algorithm, SCADE, for simultaneous sensor calibration and deformation estimation of CMs.
• To address the challenge of fusing continuous, inaccurate embedded sensor data with intermittent, accurate external measurements.
• To improve the accuracy of CM shape and position estimation in both free and obstructed surgical environments.

Main Methods:

• Developed a novel mathematical formulation for an a priori model-independent filtering algorithm.
• Integrated continuous, inaccurate embedded sensor (e.g., FBG) measurements with intermittent, accurate external imaging data (e.g., cameras).
• Validated the SCADE algorithm using experiments with a CM designed for orthopaedic interventions.

Main Results:

• SCADE successfully estimated the unknown deformation behavior of an unmodeled CM.
• The algorithm effectively identified and compensated for time-varying drift in a poorly calibrated Fiber Bragg Grating (FBG) sensing unit.
• SCADE demonstrated significantly superior performance in CM tip position estimation compared to FBG-based methods.

Conclusions:

• The SCADE algorithm provides a robust solution for accurate CM deformation and position estimation in surgical applications.
• SCADE enables simultaneous calibration of sensors and estimation of manipulator shape, enhancing robotic surgery capabilities.
• This approach offers a significant advancement over traditional methods relying solely on embedded sensor data.