top of page

TAIWAN – LATVIA –LITHUANIA COLLABORATION

臺灣、拉脫維亞、立陶宛

​國際合作

Development of Microrobot Based on Visual Recognition and Machine Learning for Manipulation of Individual Living Cells

基於視覺辨識和機械學習開發用於單一活細胞操作的微動機器人

Project Details 計畫內容概述

The project’s theme aligns with the “Information and Emerging Production Technologies research domain.” Our primary objective is to develop an innovative, autonomous, and flexible microrobot designed to precisely manipulate individual living cells. Micromanipulation systems typically suffer from two key limitations: lack of autonomy and object recognition capabilities, necessitating highly skilled operators. Our proposed system addresses these challenges comprehensively by integrating intelligence, autonomy, and automated data processing. It comprises three main components:

  1. A micro-positioning system is facilitating 3D displacement of the manipulation tool.

  2. A video recognition system for identifying micro-objects, including properties like size, shape, object boundaries, and breakpoints.

  3. ​A machine learning system for autonomous control.

 

The video recognition system empowers the system to discern object characteristics, enhancing its precision. Notably, our system is designed to function autonomously, capable of manipulating and positioning objects independently through machine learning-based algorithms. While such systems remain relatively rare due to their high design costs and substantial resource requirements, this international collaboration program presents an opportunity to harness the collective expertise of specialists from diverse scientific disciplines. Together, we can overcome these challenges and cultivate groundbreaking technology with significant potential for advancing micro-object research.

本計畫的主題與「資訊與新興生產技術研究領域」相符。我們的主要目標是開發一款創新、自主且靈活的微型機器人,旨在精確操作單個活體細胞。微操作系統通常面臨兩個關鍵限制:缺乏自主性和物體識別能力,需要具備專業技術的操作人員。我們提出的系統全面解決了這些挑戰,通過整合智慧化、自主性和自動化數據處理來實現。該系統包括三個主要部分:

  1. 微定位系統,促進操控工具的三維位移​

  2. 視訊識別系統,用於識別微小物件,包括大小、形狀、物件邊界和斷點等特性​

  3. 機器學習系統用於自主控制

視覺辨識系統使系統能夠辨識物體特徵,提高其精確度。值得注意的是,我們的系統設計為能夠自主運作,能夠透過基於機器學習的演算法獨立操控和定位物體。雖然這類系統由於高昂的設計成本和龐大的資源需求而相對罕見,但這個國際合作計畫提供了一個機會,可以整合來自不同科學學科的專家集體專業知識。我們共同努力,可以克服這些挑戰,培育具有重要潛力的突破性技術,以推動微觀物體研究的進展。

Project

Duration

2020~2023

Project No.

MOST 110-2923-E-011-006-MY3

Project Partners

  1. National Taiwan University of Science and Technology (NTUST) in Taiwan

  2. Institute of Electronics and Computer Science (EDI) in Latvia
  3. Vilnius Gediminas Technical University (Vilnius Tech) in Lithuania

Principal Investigators

  1. Professor Po Ting Lin (Taiwan)

  2. Professor Modris Greitans (Latvia)

  3. Professor Vytautas Bučinskas (Lithuania)

Contact 聯絡資訊

Publication 期刊論文發表

With the support of this project, this research team has published the following four international journal papers, and one paper is under review.

  •  Zheng-Jie Huang, Brijesh Patel, Wei-Hao Lu, Tz-Yu Yang, Wei-Cheng Tung, Vytautas Bučinskas, Modris Greitans, Yu-Wei Wu, Po Ting Lin*, Yeast Cell Detection Using Fuzzy Automatic Contrast Enhancement (FACE) and You Only Look Once (YOLO), Scientific Reports, 13, 16222, 2023.

  • Subačiūtė-Žemaitienė Jurga, Andrius Dzedzickis, Antanas Zinovičius, , Vadimas Ivinskij, , Justė Rožėnė, Rokas Bagdonas, Vytautas Bučinskas, Scanning Electrochemical Microscope Based on Visual Recognition and Machine Learning, Key Challenges in Automation, Robotics and Measurement Techniques. AUTOMATION 2023. Lecture Notes in Networks and Systems, vol 630. Springer.

  • Laizans, Modris, Janis Arents, Oskars Vismanis, Vytautas Bučinskas, Andrius Dzedzickis, and Modris Greitans. Supplementation of synthetic object replicas for increasing precision of microrobot trajectory keypoints, Robotic Systems and Applications, 3, no. 1 (2023): 47-58.

  • Chen, Yen Jung, Wei-Cheng Tung, Wei-Rui Lee, Brijesh Patel, Vytautas Bučinskas, Modris Greitans, and Po Ting Lin*. Designing and controlling a self-balancing platform mechanism based on 3-RCC spherical parallel manipulator, Robotic Systems and Applications (2023).

  • Vytautas Bucinskas, Jurga Subaciute-Zemaitiene, Andrius Dzedzickis, Inga Morkvenaite-Vilkonciene, Robotic micromanipulation: a) actuaors and their application, Robotic Systems and Applications, 1(1), pp. 2-23, 2021.

Besides, six international conference papers were published, as shown below:

  • Wei-Cheng Tung, Brijesh Patel, Vytautas Bučinskas, Andrius Dzedzickis, Modris Greitans, Po Ting Lin* (2023.8.30-9.1) Automatic Cell Recognition in Microscopic Images Based on Contrast Enhancement and YOLO, 2023 International Conference on Advanced Robotics and Intelligent Systems (ARIS 2023), Paper Number 1017, Taipei, Taiwan

  • Makulavicius, Mantas, Justas Makutėnas, Dominykas Čičiurėnas, Vytautas Bučinskas, and Andrius Dzedzickis(2023.06.20 – 06.22), The Concept of Efficient Tool Path Generation Method for Industrial Robots, International Workshop on Embedded Digital Intelligence (IWoEDI’2023), Riga, Latvia.

  • Oskars Vismanis*, Janis Arents, Vytautas Bucinskas, Andrius Dzedzickis, Brijesh Patel , Wei-Cheng Tung, Po Ting Lin and Modris Greitans(2023.06.20 – 06.22), Design and implementation of a microrobot system for manipulation of microscopic objects, International Workshop on Embedded Digital Intelligence (IWoEDI’2023), Riga, Latvia.

  • Zheng-Jie Huang, Imran Ali, Vytautas Bučinskas (Lithuanian PI), Modris Greitans (Latvian PI), Po Ting Lin* (Taiwanese PI) (2021.8.17-21) Fuzzy Automatic Contrast Enhancement (FACE) to Improve the Performance of Object Detection of Yeast Cell Images, 2021 International Conference on Advanced Robotics and Intelligent Systems (ARIS 2021), Paper Number 1129, Virtual Online, Taiwan.

  • Modris Laizans, Oskars Vismanis, Janis Arents, Vytautas Bučinskas, Po Ting Lin, Modris Greitans* (2021.8.17-21) Improvement of An Object Detectors Precision by Diversifying the Training Dataset with Image Enhancement and Synthetic Data, 2021 International Conference on Advanced Robotics and Intelligent Systems (ARIS 2021), Paper Number 1127, Virtual Online, Taiwan.

  • Morkvėnaitė-Vilkončienė, A. Dzedzickis, J. Subačiūtė-Žemaitienė, Modris Greitans, Po Ting Lin*, Vytautas Bučinskas* (2021.8.17-21) Development of Automated System Concept Intended for the Electrochemical Evaluation of Micro-Objects, 2021 International Conference on Advanced Robotics and Intelligent Systems (ARIS 2021), Paper Number 1072, Virtual Online, Taiwan.

 

Also, one Taiwanese invention patent was awarded by this research team.

  • Po Ting Lin*, Yu-Wei Wu, Wei-Hao Lu, Shu-Ping Lin, Yu-Ta Yao, Modris Greitans, Vytautas Bučinskas, “Highly adaptable image coding method and image recognition method”, Taiwan Invention Patent: I800874

bottom of page