In this paper a novel approach to learning by demonstration (LbD) is presented. A multimodal service robot is taught grasping skills by a human instructor who demonstrates a grasping action. Our approach contributes novel solutions to the aspects of robustly tracking the demonstrator's hands in real time as well as to the transformation of tracking results into grasping skills. To track the demonstrator's hands in stereoscopic images a Mean-Shift-like algorithm is adapted. For the very first time this algorithm is applied to local binary patterns (LBP) and color histograms. To retrieve the hand configuration we use view-based Principal Component Analysis (PCA). To develop grasping skills from tracking results the robot repetitively tracks the demonstrator's grasping actions and transforms the results into three-dimensional self organizing maps (SOMs). The SOMs give a spatial description of the collected data and serve as data structures for a reinforcement learning (RL) algorithm which optimizes trajectories for use by the robot. The approach is applied to a multimodal service robot. Experiments show the effectiveness of the LBP-enhanced Mean-Shift-like tracking and the robustness of LbD based on SOMs and RL

No abstract

In this paper we propose a novel concept for the programming of multi-modal service robots. The presented software architecture eases the development of high-level applications for service robots. The software architecture is based upon the Roblet-Technology, which is a powerful medium for robots. It introduces the possibility to develop, compile and execute an application on one workstation. Since the Roblet-Technology uses Java the development is independent of the operation system. With the feature of running programs as a distributed software, the framework allows running algorithms which need great computation power on different machines which provide this power. In this way, it greatly improves programming and testing of applications in service robotics. The concept is evaluated in the context of the service robot TASER of the TAMS Institute at the University of Hamburg. This robot consists of a mobile platform with two manipulators equipped with artificial hands. Several multimodal input and output devices for interaction round off the robot.

No abstract

No abstract