Adaptive path-following control for bio-inspired steerable needles

Abstract

Needle steering systems have shown potential advantages in minimally invasive surgery in soft-tissue due to their ability to reach deep-seated targets while avoiding obstacles. In general, the control strategies employed to drive the insertion use simplified kinematic models, providing limited control of the trajectory between an entry site and a deep seated target in cases of unmodelled tissue-needle dynamics. In this work, we present the first Adaptive Path-Following (APF) controller for a bio-inspired multi-part needle, able to steer along threedimensional (3D) paths within a compliant medium by means of the cyclical motion of interlocked segments and without the need for duty-cycle spinning along the insertion axis. The control strategy is outlined in two parts: a high-level controller, which provides driving commands to follow a predefined 3D path smoothly; and a low-level controller, able to counteract unmodelled tissue-needle nonlinearities and kinematic model uncertainties. A simulation that mimics the needle's mechanical behavior during insertion is achieved by using an Experimental Fitting Model (EFM), obtained from previous experimental trials. The Simulation results demonstrate the robustness and adaptability of the proposed control strategy.

Cyclic Motion Control for Programmable Bevel-Tip Needle 3D Steering: A Simulation Study

“…A 3D linear inverse model, presented in [17] and based on an unmanned, aerial vehicle, is used to approximate the relationship between desired curvature and segment offsets via an optimization function. These segment offsets are then sent to both a low level cyclic controller which determines the velocity and displacement profiles for the motor to follow, as well as a low level direct controller which determines the velocity and displacement profiles when no cyclic motion is applied.…”

Modelling the Deformation of Biologically Inspired Flexible Structures for Needle Steering

“…Frasson et al [4] subsequently demonstrated that a 2-segment needle design could perform 2D steering through a soft substrate with a controllable curvature, and later Secoli et al demonstrated that 3D steering was achievable with a 4-segment needle. A nonlinear adaptive control strategy was implemented, based on a linear model of the needle, and shown to guide the needle along 3D trajectories [5]. With the increase in number of segments, and complexity of the design, the steering behaviour was no longer well described by the linear model, which motivates the development of a mechanics-based model for the deformation of these unique structures.…”

A Mechanics-Based Model for 3-D Steering of Programmable Bevel-Tip Needles

“…To date, the control strategy for PBNs has assumed a linear model with adaptive compensation for unknown nonlinearities [25], [26]. However, to improve existing control strategies and inform future design decisions, the highly nonlinear relationship between the needle tip shape and steering needs to be better understood.…”