Neural Network Control of Optical Tweezers System for Manipulation of Microscopic Objects

Abstract

Though different techniques have been formulated in the past for the optical micro-manipulation, the feasibility of these techniques mostly relied on the common assumption of the known structure of robotic tweezers dynamics. However, in most cases, the system has unmodeled dynamics because of which it is difficult to comprehend the structure of the regressor matrix. This creates complications in the designing and implementation of controllers for the optical tweezers systems. In this paper, we propose a neural network-based controller for set-point control of an optical tweezers system with uncertain dynamics. We use the neural networks to approximate the dynamics of the robotic tweezers and thus the proposed method allows the control of the system without knowing the structure of the dynamic model. Numerical simulations are also presented to demonstrate the effectiveness of the proposed approach. Index Terms—Optical tweezers, Cell manipulation, Neural Network, Set-point control.