Alexa Hauck

Learn More
In the eld of visual control of robot motion, visual servoing has been proposed as the suitable control strategy to cope with imprecise models and calibration errors. Remaining problems such as the necessity of a high rate of visual feedback are deemed to be solvable by the development of real-time vision modules. However, human grasping, which still(More)
This paper presents an approach to the recognition of articulated 3D objects in monocular video images. A hierarchical object representation models objects as a composition of rigid components which are explicitly connected by speciic kinematic constraints, e.g., rotational and/or translational joints. The recognition task follows this tree-like structure(More)
We present an object recognition system that identifies objects by their silhouette from single views of a greyscale camera. The centroidal profile describing the object boundary is matched with boundaries from a model base using a dynamic programming technique. Objects are modelled by a multi-view representation which can either be learned from a set of(More)
The design of mobile robots which can cope with unexpected disturbances like obstacles or misplaced objects is an active eld of research. Such an autonomous robot assesses the situation by comparing data from one or more sensors with an internal representation of its environment. In this paper we present a hierarchically structured world model that combines(More)
Human grasping still outshines its robotical counterparts with respect to accuracy, speed, robustness, and flexibility. When trying to develop a robotical hand-eye system, it is therefore only natural to examine the results of neuroscience. In this paper, we examine the human hand-eye system concerning motion planning and control using robotical categories(More)
Pose estimation is a classical problem in computer vision applications like object recognition or robot navigation. Most approaches deal with rigid objects only, though. We propose an algorithm for the determination of joint states that is based upon a hierarchical object model. This allows to break down the complex registration task into simpler ones, as(More)
Perception tasks for autonomous mobile robots such as navigation, object-and state-identiication require reliable and accurate model information as complete as possible. Early disposal of a closed and prevailingly static description of the environment improves the performance of these tasks. In opposite to these requirements, exploration preliminary(More)