Ellipse Detection-Based Bin-Picking Visual Servoing System

Ellipse Detection-Based Bin-Picking Visual Servoing System

Kai Liu (Tsinghua University, China), Hongbo Li (Tsinghua University, China) and Zengqi Sun (Tsinghua University, China)
Copyright: © 2013 |Pages: 8
DOI: 10.4018/978-1-4666-4225-6.ch008


In this chapter, the authors tackle the task of picking parts from a bin (bin-picking task), employing a 6-DOF manipulator on which a single hand-eye camera is mounted. The parts are some cylinders randomly stacked in the bin. A Quasi-Random Sample Consensus (Quasi-RANSAC) ellipse detection algorithm is developed to recognize the target objects. Then the detected targets’ position and posture are estimated utilizing camera’s pin-hole model in conjunction with target’s geometric model. After that, the target, which is the easiest one to pick for the manipulator, is selected from multi-detected results and tracked while the manipulator approaches it along a collision-free path, which is calculated in work space. At last, the detection accuracy and run-time performance of the Quasi-RANSAC algorithm is presented, and the final position of the end-effecter is measured to describe the accuracy of the proposed bin-picking visual servoing system.
Chapter Preview

Quasi-Ransac Ellipse Detection

Since we choose cylinder as target object, and the projection of cylinders in image are some ellipses, so the target detection task is boiled down to ellipse detection task. The detection result is organized as

(1) where 978-1-4666-4225-6.ch008.m02 denote the coordinates of ellipse center in image, 978-1-4666-4225-6.ch008.m03are its semi-major axis and semi-minor axis, and 978-1-4666-4225-6.ch008.m04 denotes its orientation angle. They are all the information about the target that the robot controller can utilize. The ellipse detection algorithm is based on the continuous edge feature which is called a contour (Cai, Yu, & Wang, 2004). In this way, we only regard the contour as a candidate of ellipse rather than all the edges in the edge image which cost lots of computation and memory.

Complete Chapter List

Search this Book: