Virtual environment as a tool for analysing the operation of an industrial robot

Jakubiec, Beata

doi:10.15199/48.2020.02.23

Cited by 4 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nowadays, industrial robots are also developing towards intelligence. Mostly, the robot is combined with visual technology, enabling them to get the characteristics of these target objects, classify and identify them in the process of parts packaging, disassembly and transportation [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Application of Robot Sorting Technology in Automatic Sorting of 3d Vision Target Pose Manipulator

2021

IJOMAM

View full text Add to dashboard Cite

A disordered sorting platform is designed to realize the application of robot sorting technology in the automatic sorting of 3D vision target positioning manipulator. First, the camera is used to detect the object to be sorted visually in order to obtain the pose information of the object. According to the environment and the specific information related to the Salsa object, the hardware of the sorting system is selected, and the process and scheme of the whole system are designed. Second, the calibration of the robot vision system composed of Kinect camera and ABB1200 robot is studied. The connection between the Kinect camera and the robot is determined as the installation mode of Eye-to-hand, and the Kinect camera itself is calibrated. Then, the Eye-to-hand system composed of the robot and the camera is calibrated, and the position relationship between the camera coordinate system and the robot base coordinate system is obtained. Subsequently, the pose estimation of randomly stacked objects is studied. The point cloud information collected by the camera is pre-processed, and the template is preliminarily matched with the target object by using the SHOT (Signature of Histogram of Orientation). And then the ICP (Iterative Closure Point) algorithm is used for further point cloud registration. Finally, the pose of the target object is estimated under the Eye-to-hand calibration results. The experimental platform of the whole unordered sorting system is built, and the multiple unordered sorting experiments of multiple objects are carried out, followed by the analysis of the experimental results. The error between the pose estimation results and the actual position is about 4mm, which verifies the feasibility of the system scheme and method and enables the robot to perform the unordered sorting task in centimetres, meeting the design requirements of the system. Therefore, the study provides a reference for future related research.

show abstract

Section: Introductionmentioning

confidence: 99%