Bin-Picking : New Approaches for a Classical Problem

In this book threedifferent approaches to solve the bin-picking problem are described, showinghow modern sensors can be used for efficient bin-picking as well as how classicsensor concepts can be applied for novel bin-picking techniques.

This bookis devoted to one of the most famous examples of automation handling tasks –the “bin-picking” problem. To pick up objects, scrambled in a box is aneasy task for humans, but its automation is very complex. In this book threedifferent approaches to solve the bin-picking problem are described, showinghow modern sensors can be used for efficient bin-picking as well as how classicsensor concepts can be applied for novel bin-picking techniques. 3D pointclouds are firstly used as basis, employing the known Random Sample Matchingalgorithm paired with a very efficient depth map based collision avoidancemechanism resulting in a very robust bin-picking approach. Reducing thecomplexity of the sensor data, all computations are then done on depth maps. This allows the use of 2D image analysis techniques to fulfill the tasks andresults in real time data analysis. Combined with force/torque and accelerationsensors, a near time optimal bin-picking system emerges. Lastly, surfacenormalmaps are employed as a basis for pose estimation. In contrast to knownapproaches, the normal maps are not used for 3D data computation but directlyfor the object localization problem, enabling the application of a new class ofsensors for bin-picking.