An Integrator-Backstepping Control Approach for Three-Dimensional Needle Steering

Adaptive energy shaping control of robotic needle insertion

“…The maximum deviation between trials at insertion completed (time = 3 seconds for illustrative purposes) is 0.1 mm for 1 and 0.5 degrees for 2 with Controller 1, and 1.9 mm for 1 and 0.9 degrees for 2 with Controller 2. The control laws (11) and (17) and the adaptive law (3) have been implemented on an embedded microcontroller (mbed NHP LPC1768) mounted on a specially designed PCB. The microcontroller provides a pressure setpoint to the valves and acquires data from the force sensor and from the optical encoder.…”

“…Proposition 2: Consider the system (2) under Assumptions 1 and 2 in closed-loop with the control law (11). In particular, assume < / , ′ = 0, and |̇| < (̇) 2 for some > 0.…”

“…Further simulation results with different parameters that correspond to the experimental setup are presented in Section 4.2. The tuning parameters are = 2; = 15 and = 50 for Controller 1, denoting the control law (11) The simulation results in Figure 3 show that the Controller 1 drives the needle to the desired depth rapidly but with a noticeable tip rotation. The insertion speed is approximately 2 mm/s, which is comparatively low in clinical practice.…”

“…This approach can be used for the purpose of needle steering, thus achieving curved insertion paths to avoid obstacles. Conversely, following a straight insertion path requires a constant axial rotation, which can damage soft tissues [11]. To overcome this problem different strategies have been devised, including: a programmable needle tip consisting of multiple segments pushed independently at the back [12]; a multi-segment needle consisting of nitinol wires [13]; a steerable needle actuated by heat transmitted with an optic fibre [14]; concentric tube robots used as steerable needles [15].…”

“…In spite of these remarkable advances, conventional metallic needles remain ubiquitous in clinical practice, thus the study of control strategies for the insertion of standard needles is an active area of research [16]. Important results in this sense include: sliding-mode control algorithms [17,18]; modelbased optimal control of flexible needles [19,20]; an optimized path planning procedure for insertions in multi layered tissues [21]; an adaptive output-feedback controller [22]; an integrator backstepping approach to reduce needle deflection in 3D [11]. Notably, these approaches aim to influence the needle deflection as the insertion progresses by means of axial rotation of bevel-tip needles.…”