Fast visual part inspection for bin picking

In this paper we present a novel 3D sensing approach for industrial bin-picking applications that is low-cost, fast and precise. The system uses a simple laser-line emitter. While a robot arm moves the object through the laser light, a synchronized camera captures the laser line image on the object. A full point cloud as well as an edge point cloud suitable for subsequent pose estimation is generated by the developed system. The aim of this work is to deliver an accurate point cloud based on which an object pose can be generated to support a manufacturing robot to deliver an object with high precision. The experimental evaluation of our system shows robust and accurate scanning results.

[1]  René Dencker Eriksen,et al.  3D scanning of object surfaces using structured light and a single camera image , 2011, 2011 IEEE International Conference on Automation Science and Engineering.

[2]  Saeid Nahavandi,et al.  Measuring depth accuracy in RGBD cameras , 2013, 2013, 7th International Conference on Signal Processing and Communication Systems (ICSPCS).

[3]  Akio Kosaka,et al.  Vision-based bin-picking: recognition and localization of multiple complex objects using simple visual cues , 1996, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. IROS '96.

[4]  Kourosh Khoshelham,et al.  Accuracy analysis of kinect depth data , 2012 .

[5]  J. Navarro-Pedreño.  Numerical Methods for Least Squares Problems , 1996 .