Structural Design and Motion Control of Magnetically Driven Microrobots

Abstract

Untethered microrobots have demonstrated their great potential in various microscale biomedical applications. Among many non-contact driving methods for microrobots, magnetic field-based actuation has been widely used due to its advantages of strong driving ability, precise control, and high safety. However, navigation control of magnetic microrobots with high accuracy and effectiveness remains several key challenges, hindering the successful application of this promising technology. These challenges are originally from capability limitations of electromagnetic actuation systems, the imprecise system modeling and complex biological application environment with multifarious uncertainties. In addition, the application scenarios of existing microrobots are mainly limited in liquid-filled environments. Therefore, this thesis aims to improve the motion ability and scene applicability of magnetic microrobots based on optimization of electromagnetic actuation system, the system model compensation, robust control strategy, and new structure design of microrobots. This study is performed in the three following perspectives.

First, the electromagnetic drive systems based on visual feedback required for the follow-up of this study were established. The theoretical model analysis of the commonly used gradient magnetic field and rotating magnetic field is carried out. On this basis, the shape of the iron cores of the electromagnetic system are optimized to obtain a larger magnetic field driving force and effective workspace. In addition, the relevant control parameters of the Helmholtz coil system are analyzed. The microrobot is extracted by image processing technology to obtain the three dimensional(3D) environment map and the real-time position and velocity information of the microrobot movement. This lays a solid foundation for the automatic control of microrobots.

Second, this research work proposes an approach that utilizes a gradient field-based electromagnetic coil system to automatically manipulate the microrobot for motion control. On the basis of the microrobot dynamics modeling, Prandtl-Ishlinskii (P-I) model describing the hysteresis characteristics of the system is introduced to establish the dynamics model of the microrobot considering hysteresis. The breadth-first search (BFS) algorithm in combination with a genetic algorithm (GA)-based local optimal trajectory planner is designed while considering the vessel network constraints and trajectory optimization. An adaptive sliding mode controller equipped with a nonlinear disturbance observer is developed for navigating the microrobot in the endovascular environment. The adaptive control law is derived from the Lyapunov stability theorem. The nonlinear disturbance observer is constructed to provide estimations of unknown disturbances and feedforward compensations for control amounts. A model-free backstepping sliding mode controller without hysteresis compensation is developed for comparison with the proposed method. Both in vitro and in vivo experiments are conducted to demonstrate the effectiveness of the proposed control approach successfully.

Finally, aiming at the problem of limited working efficiency and applicable environment of a single microrobot driven by the gradient magnetic field, a bioinspired magnetically driven microwalker based on frictional anisotropy is proposed. This microwalker consists of two rigid segments with an equal length of 70µm connected by a rigid joint and is fabricated integrally by 3D laser lithography without further assembly, thereby simplifying microrobot fabrication while enhancing structural integrity. Parallel setae-like tentacles are placed at the bottom of the segments as contact feet to generate friction with the contact surface. The various parameters of the microwalker are optimized through motion analysis and experimental verification. Under an external oscillating magnetic field, the microwalker can be well controlled to move forward in the low Reynolds (Re) number regime. Different from the existing microswimmers, the microwalker can achieve surface motion in non-liquid filled environments including climbing the slope.

In summary, the established electromagnetic actuation system based on visual feedback can lay the foundation for the motion control of microrobots, the proposed control method with hysteresis compensation can solve the microrobot control problem in complex microenvironment to improve motion accuracy of microrobots, and the designed microwalker with new structure expands the application scenarios of microrobots. The research of this thesis will potentially benefit the further development of microscale biomedical applications.

Date of Award	27 Mar 2023
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Dong SUN (Supervisor) & Yong WANG (External Supervisor)