Automated Three-dimensional Manipulation of Magnetic Microrobots

磁性微型機器人的自動化三維操作

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
  • Haibo Ji (External person) (External Supervisor)
  • Dong SUN (Supervisor)
Award date14 Mar 2022

Abstract

Microrobots and functional microparticles have exhibited great prospects in biomedical engineering. In most of their applications, microrobots must move from the initial position to the target position to complete their tasks. However, the automatic manipulation of microrobots in broad three-dimensional (3D) space remains a huge challenge due to the small size of these robots and the lack of an effective real­time imaging technology for locating them. Moreover, the carrying capacity of a single microrobot is considerably limited by its volume constraint, which also increases the difficulty of detection and manipulation. A microrobotic swarm composed of numerous magnetic microrobots or microparticles can effectively improve delivery efficiency while retaining flexibility and adaptability. Thus, it is considered a promising manipulation scheme. In this thesis, a field-of-view tracking scheme is proposed to achieve the 3D navigation control of microrobots in centimeter-scale range, and a magnetic tweezer system is designed to realize the surface-assisted 3D navigation of microrobotic swarms in complex microenvironments. This thesis includes three major parts.

First, a systematic approach that uses a six­-coil electromagnetic system is presented to realize automated 3D mapping and path planning in a large workspace. Two movable orthogonal microscopic cameras automatically scan the microenvironment from the top and side views. A shake correction vector is introduced to correct the original images, reducing random errors in the map caused by mechanical vibration. The obstacle area and accessible area are identified through the adaptive binarization processing of scanning sequences, and a stereo occupancy map is constructed. Moreover, the initial position of the microrobot can be located on the basis of the similarity curves generated by the scanning sequences. In accordance with the distribution characteristic of electromagnetic fields in the workspace, an enhanced rapidly exploring random tree algorithm is proposed to avoid obstacles in complex environments.

Second, a field-of-view tracking scheme is designed using a specific image definition evaluation algorithm to ensure continuous visual servo in a wide range. A prescribed performance controller with a disturbance observer is designed in according with the dynamic model of the camera's driving stage. This controller guarantees that the microrobot can always remain within the view of microscopic cameras. Accordingly, high-quality feedback images can be obtained for the position estimation and motion control of the microrobot. Another prescribed performance controller is designed for the microrobot motion to enable its transient and steady-state performance to satisfy the expected requirements. Simulations and experiments are performed to verify the effectiveness of the proposed automatic 3D navigation strategy. Experimental results show that a microrobot can navigate automatically in a centimeter-scale microenvironment with obstacles, and achieve reliable field-of-view tracking and path following at a high speed.

Third, a novel magnetic tweezer system with a large workspace is developed to manipulate microswarms efficiently. The magnetic tweezers generate rotating magnetic fields in the workspace, enabling magnetized microparticles to roll toward a specific point along spiral trajectories. The assembly mechanism of a microswarm under magnetic tweezers is analyzed and verified. The developed system can assemble low-density magnetic microparticles to form a stable and compact swarm at a predetermined position, and the swarm position can be precisely controlled by driving the magnetic tweezers without relying on real-time image feedback. Because the flow rate near the channel wall is much lower than that in the channel center, microswarms can maintain stable aggregation and controlled movement in a lower magnetic field. In addition, a microswarm can be rapidly disassembled into microchains and uniformly spread in a large range with the aid of the out-of-step behavior of microparticles. The experimental results demonstrate that the microswarm exhibits satisfactory motion performance and delivery efficiency.

In summary, this study presents feasible automated 3D manipulation schemes for microrobots and microswarms that enable them to precisely navigate in microenvironments. It provides a meaningful reference for the future design and application of various microrobots and functional microparticles.