Publications

everal animal species like rats or seals explore the surrounding environment with their sensory hairs, the so-called vibrissae. Regarding to natural vibrissa, the functionalities of tactile object sensing to receive information about, e.g., the shape and/or the surface texture are often discussed. But, the usage of natural vibrissae to detect flows is reported as well. The necessary information about the flow are coded in the signals recorded at the base of the vibrissa. Due to the natural paragon, we adapt this behavior in order to show up how the information about flows can be extracted from these observables. This is an important procedure because the detection and determination of information about flow effects become interesting for several reasons. For example, in the case of other sensing tasks and strategies like object contour scanning, flows are considered as (possible) disturbances that have to be known to reconstruct any sensed boundary. Otherwise, the flow itself can be of interest. In a first step, simulations are performed to examine the relations between the observables at the base of the sensor and the applied velocity of a flow. Here, a steady-state uniform flow is assumed. The simulations are validated by experiments.

Der Gegenstand der Arbeit sind Tensegrity-Strukturen mit mehreren stabilen Gleichgewichtskonfigurationen, sogenannte multistabile Tensegrity-Strukturen. Im Vordergrund der Arbeit steht die Entwicklung von Algorithmen, mit denen solche Strukturen entworfen, untersucht und gezielt ausgelegt werden können. Dafür werden Möglichkeiten zur Bestimmung der Gleichgewichtskonfigurationen von multistabilen Tensegrity-Strukturen betrachtet. Des Weiteren wird untersucht, wie Tensegrity-Strukturen so ausgelegt werden können, dass sie vorgegebene Eigenschaften aufweisen. Dazu werden Kenngrößen zur Charakterisierung dieser Eigenschaften definiert. Für beide Aufgabenstellungen werden Optimierungsprobleme hergeleitet. Zur Lösung dieser Optimierungsprobleme werden Algorithmen entworfen, getestet und analysiert. Aufbauend auf diesen theoretischen Untersuchungen liegt ein weiterer Schwerpunkt dieser Arbeit in der Betrachtung der Einsatzmöglichkeiten von multistabilen Tensegrity-Strukturen in der Greifertechnik. Es werden verschiedene Konzepte für die Entwicklung von Greifern aus diesen Strukturen diskutiert. Zu ausgewählten Konzepten erfolgen weiterführende Betrachtungen, unter anderem durch Einbeziehung dynamischer Analysen. Neben theoretischen Untersuchungen dieser Greifer werden die wichtigsten Erkenntnisse experimentell an Funktionsmustern überprüft und potentielle Einsatzgebiete werden aufgezeigt.

For many practical applications it is important to determine not only a numerical approximation of one but a representation of the whole set of globally optimal solutions of a non-convex optimization problem. Then one element of this representation may be chosen based on additional information which cannot be formulated as a mathematical function or within a hierarchical problem formulation. We present such an application in the field of robotic design. This application problem can be modeled as a smooth box-constrained optimization problem. We extend the well-known alphaBB method such that it can be used to find an approximation of the set of globally optimal solutions with a predefined quality. We illustrate the properties and give a proof for the finiteness and correctness of our modified alphaBB method.

Rodents, like mice and rats, use tactile hairs in the snout region (mystacial vibrissae) to acquire information about their environment, e.g. the shape or contour of obstacles. For this, the vibrissa is used for the perception of stimuli due to an object contact. Mechanoreceptors are processing units of this stimuli measured in the compliant support (follicle sinus complex). We use this behavior from biology as an inspiration to set up a mechanical model for object contour scanning. An elastic bending rod interacts with a rigid obstacle in the plane. Analyzing only one quasi-static sweep of the rod along the obstacle (in contrast to literature), we determine a) the support reactions (the only observables of the problem), and then b) the (discrete) obstacle contour in form of a set of contact points. In doing this, we first assume a stiff support (clamping) of the vibrissa, but in a next step we increase the elasticity of the support in focussing on a bearing with a rotational spring (also to control or delimitate the bending moment at the support). Thereby, we present a fully analytical treatment of the non-linear differential equations emerging from Bernoulli's rod theory and a representation by Standard Elliptic Integrals.