Computer vision system R&D for EAST Articulated Maintenance Arm robot
| Autor: | Yang Yang, Hansheng Feng, Ling-Long Lin, Yuntao Song, Pan Hongtao, Yong Cheng |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Computer science
business.industry Mechanical Engineering ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Base (geometry) Robotics Edge detection Superconducting tokamak Identification (information) Nuclear Energy and Engineering EamA Robot General Materials Science Computer vision Artificial intelligence business Invariant (computer science) Civil and Structural Engineering |
| Zdroj: | Fusion Engineering and Design. 100:254-259 |
| ISSN: | 0920-3796 |
| DOI: | 10.1016/j.fusengdes.2015.06.017 |
| Popis: | Experimental Advanced Superconducting Tokamak (EAST) is the first full superconducting tokamak device which was constructed at Institute of Plasma Physics Chinese Academy of Sciences (ASIPP). The EAST Articulated Maintenance Arm (EAMA) robot provides the means of the in-vessel maintenance such as inspection and picking up the fragments of first wall. This paper presents a method to identify and locate the fragments semi-automatically by using the computer vision. The use of computer vision in identification and location faces some difficult challenges such as shadows, poor contrast, low illumination level, less texture and so on. The method developed in this paper enables credible identification of objects with shadows through invariant image and edge detection. The proposed algorithms are validated through our ASIPP robotics and computer vision platform (ARVP). The results show that the method can provide a 3D pose with reference to robot base so that objects with different shapes and size can be picked up successfully. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect