Motion planning for robot-aided optical manipulation of biological cells

Abstract

Optical tweezers exhibits high accuracy in flexible and noninvasive manipulation of microparticles. Increasing demand for both efficiency and productivity in the manipulation of biological cells highlights the need for the advanced use of optical tweezers. Recent efforts in integrating robotics into optical tweezers to create a new cell manipulation tool have drawn considerable attention in the fields of robotics and bioengineering. Transporting the trapped cells to the user-defined goal region while avoiding collisions with other cells and obstacles present in the workspace, is a typical motion-planning problem. How to extend a robotic motion planner is important in finding a collision-free path/trajectory for transferring the trapped cells to the desired goal location, in which minimizing optical damage to the trapped cells should also be considered. This thesis aims to develop a motion planning method for cell transportation with collision avoidance and minimum optical damage to the trapped cells. The research is carried out in the following three perspectives. First, a new approach to integrating a robotic path planner into an optical tweezers manipulation system is developed in order to achieve automated transportation of live cells. A Rapidly Exploring Random Tree (RRT) based path planner is applied for the first time to the cell transportation application. Experiments on transferring the yeast cells are performed to demonstrate the effectiveness of RRT-based path planning especially in stable aqueous solution. Second, a dynamic path planner with an online monitoring strategy is further developed in dynamic solution to avoid collisions dynamically, in which the environmental influence caused by the Brownian movement of the other cells is particularly taken into account. The proposed dynamic path planner can successfully deal with the complex dynamic environments. Furthermore, the proposed path planner shows potential for application in 3D cell transportation by dividing 3D cell transportation tasks into two sub-tasks in two orthogonal 2D planes. Third, an energy model for optically trapped cells is established for the objective of minimizing the optical damage of trapped cells during the cell transportation. In path planning, an A* based path planner is used to design a collision-free path. In trajectory smoothing along the generated path, the trajectory is further optimized through parameter optimization with a new objective function that considers the minimum optical damage of trapped cell based on the established energy model. This thesis study makes an important contribution to the illustration of using a robotic path planner to address the automated transportation of biological cells with laser setup in both stable and dynamic aqueous solutions. The issue of minimizing optical damage to the trapped cell caused by laser radiation is also considered in the path planning.

Date of Award	15 Jul 2013
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Dong SUN (Supervisor)