Relevant Publication

• S. Raval, O. Erin, X. Liu, L. Mair, W. Pryor, Y. Barnoy, I. Weinberg, A. Krieger, Y. Diaz-Mercado, “Magnetic Model Calibration for Tetherless Surgical Needle Manipulation using Zernike Polynomial Fitting”, in IEEE 21st International Conference on Bioinformatics and Bioengineering (BIBE), 2021, pp. 1-6. [Paper]

For magnetic robot manipulation near coils, commonly employed magnetic dipole models are not accurate enough to provide precise control. A complete estimation of the forces and torques on the magnet can be achieved by finite element models (FEM), which consider both the electromagnet and the needle magnetization, volume, and shape. However, such computations do not provide differentiable closed-form expressions. To find a solution to the nonlinear magnetic field distribution that is sufficiently fast and efficient to enable safe, precise, and autonomous control of surgical magnetic robots, we approximate the nonlinear magnetic field by a linear combination of nonlinear basis functions, specifically the Zernike polynomials.

Dipole RMS Tracking Error: 2.35 mm (~10% of Needle Length)