Testing the detection and recognition of objects

Recognizing and manipulating objects is an important task for mobile robots performing useful services in everyday environments. While existing techniques for object recognition related to manipulation provide very good results even for noisy and incomplete data, they are typically trained using data generated in an offline process.Specifically, we develop an approach to building a 3D surface model of an unknown object based on data collected by a kinect sensor observing the robot’s hand moving the object.