Accurate Shape and Tip-Contact-Force Estimation of Multisegment Continuum-Robotic Tubular Bronchoscopes

Abstract

Continuum robots have found extensive applications in minimally invasive surgical interventions through the natural orifices of the human body. Due to limited application scenarios, many conventional sensors are incompatible with small-sized continuum bronchoscopes, posing challenges for tip force measurement. Therefore, a method for estimating tip force based on the tension in the drive wires and the shape of the continuum has been proposed, with control constraints designed based on the estimated forces. A linearly decreasing weights particle swarm optimization has been employed for the efficient and precise determination of shape reconstruction using two-segment Bézier curves. An experimental setup for shape reconstruction and force perception was established, and experiments were conducted on both the free state and isolated porcine lung tissue. The results demonstrate that within a bending angle of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$180^{\circ }$</tex-math></inline-formula>, the average root-mean-square error (RMSE) is 0.28 mm, which is 0.05% of the continuum length. The average RMSE of force perception is less than 10 mN. The accuracy is equivalent to the state-of-the-art continuum robot force sensing methods without the passive segment. Detailed modeling and validation results verify the accuracy and reliability of the proposed method for intracavitary interventions with microscale continuum robots.