Closed-loop vision-assisted actuation of a bubble microrobot

Abstract

Precise control and point-to-point actuation of opto-thermocapillary flow addressed (OFB) bubble microrobots are critical for applications in micromanipulation. Here, we have developed and demonstrated a hybrid closed-loop control system capable of automated point-to-point actuation of a bubble microrobot in linear trajectories, and semi-automated vision-based position updates of a bubble microrobot for precise positioning. The microrobot was actuated point-to-point using computer generated holographic (CGH) control system along a random path. At each waypoint of the trajectory the microrobot’s location within the workspace was detected from image processing by a particle tracker application developed in MATLAB and compared with the location in the CGH user interface. Subsequently, the microrobot was actuated to minimize any difference in location measured by the tracking application, enabling vision-based closed-loop control of the microrobot.