RoVam
"Development of microrobot based on visual recognition and machine learning for manipulation of individual living cells"
Status: Concluded, 2023
Tech Stack: C++, Python, PyTorch
A project of the Latvian-Lithuanian-Taiwanese scientific cooperation support fund in the research domain of “Information and new production technology.” The aim of this project was to create a new type of autonomous, flexible microrobot for manipulating individual living cells, as existing micromanipulation systems were usually not autonomous and did not have object recognition capabilities, meaning that only highly skilled personnel could operate them. The developed technology consists of three main components: a 3D micro-positioning system, a video recognition system for micro-object recognition, and a machine learning system for control.
Funding
Research Council of Lithuania (LMTLT), No. P-LLT-21-6, State Education Development Agency of Latvia, No. LV-LT-TW/2021/8, and Ministry of Science and Technology (MOST) of Taiwan, No. MOST 110-2923-E-011-006-MY3
Contribution
The main role of the robotics Group was to develop and implement a recognition system capable of detecting biological objects in small solution samples, determining their center coordinates and pick-up points by defining object boundaries and shapes, and to implement a calibration technique for aligning the coordinate systems of the imaging and mechanical manipulator systems.