Filters

Settings

Now showing1 - 8 of 8
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Dezentrale kooperative Exploration von Umweltfeldern mit kleinen autonomen Unterwasserrobotern in räumlich beschränkten Umgebungen
    Exploration and monitoring of environmental and potential hazardous fields in underwater environments, such as chemical concentration or radiation, are among the most promising tasks to be performed by groups of autonomous underwater vehicles (AUVs). While the first phase of this project concentrated on teams of AUVs in structured and fully known environments, e.g. an obstacle-free water basin, the second phase extends this work to confined and cluttered scenarios. To date, little work has examined the challenges that arise when moving from operation in controlled research environments to field exploration in real-world cluttered scenarios. Examples range from harbor basins to industrial facilities such as legacy nuclear decommissioning ponds. The main objective of this research project is the decentralized exploration and monitoring of environmental fields with teams of small-scale underwater robots. Thereby, the team of micro-AUVs coordinates their motion to estimate an unknown concentration field within the fluid volume of interest. The environmental field is modeled using a stochastic field belief representation, namely non-stationary Gaussian Markov Random Fields (GMRFs). Within this setting, robot agents acting as mobile sensor nodes compute information-rich paths that reduce the uncertainty of the estimated field effectively. Due to the massive limitations with respect to communication bandwidth in the underwater domain, approaches using a central coordinating unit render infeasible. As a consequence, decentralized concepts are investigated that rely on computationally lightweight field estimation and motion planning methods that allow implementation onboard the individual micro underwater robots. In these cluttered environments, obstacles such as walls induce anomalies in the field structure that are a major challenge and need to be included within the field belief and exploration framework. Since the unknown fields are modeled by a nonparametric approach (GMRFs) a minimum of a priori knowledge is required. The initially unknown covariance function can be estimated along the field. The approach requires analyzing a fundamental tradeoff in explorative problems summarized by the following question: Shall future observations be collected to improve the field model (a better covariance function estimate), or shall observations be taken at locations where the current (possibly inaccurate) model indicates the highest uncertainty? In the first phase of the project, a computationally lightweight field belief representation, based on GMRFs has been derived. Two planning and control strategies for autonomous exploration of environmental fields have been developed. These preliminary results are the starting point for extending a truly decentralized exploration framework in confined and cluttered environments. A thorough analysis and comparison of these strategies will be performed, and the concept will be verified experimentally.
      23
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Real-time capable model inversion using servo-constraints for fast moving soft robots
    The aim of the project is real-time capable model inversion for feedforward control of fast-moving soft robots. While the first phase of the project considered real-time model inversion for general underactuated mechanical systems, the second phase will focus on the application-specific challenges in the area of fast-moving soft robots. Soft robots are modern robots made of very soft materials, such as silicone or other elastomers, and represent a highly relevant area of research. Due to their soft structure, soft robots are particularly suitable for gripping processes, as well as for use in environments where collisions with objects or humans may occur. So far, primarily feedforward and feedback control methods based on quasi-static models have been used for soft robots. However, these approaches are not applicable to fast-moving soft robots because their dynamic behavior plays a dominant role. Therefore, the inverse models developed in this project explicitly account for the system dynamics. For these underactuated systems, key questions arise regarding their structural properties and how, based on these, real-time model inversion can be efficiently implemented for larger models with many degrees of freedom and multiple actuators. Furthermore, models for soft robots are often inaccurate due to manufacturing imperfections and parameter uncertainties. Therefore, a data-driven adaptation of the models is also pursued, ensuring compatibility with model inversion. To incorporate real parameters at an early stage and validate the developed methods, numerical studies will be closely integrated with experimental work. For this purpose, a family of soft robots is being developed, providing systems of varying complexity. These include both the inherent underactuation due to the soft nature of the robots and cases of redundant actuation that enable complex movements. The practicality of model inversion using servo constraints for various classes of underactuated multibody systems has been demonstrated in the project.
      21
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Efficient simulation for analysis and optimization of a wave energy converter in non-linear random sea states
    Obtaining energy from a random source is a fundamental problem in the field of renewable energy. In addition to the possibility of obtaining solar or wind energy, there is also the possibility of obtaining energy from ocean waves. Thereby, mechanical systems are positioned on the sea surface and electrical generators convert the motions of the systems into electrical energy. Since wave energy has a notably high power density compared to wind and solar energy, it is a promising resource for renewable energy. However, in order to harvest as much energy as possible from wave energy converters (WECs), the mechanical system has to be designed in such a way that a large dynamic motion is achieved. On real oceans, sea waves occur randomly. Furthermore, measurements have shown that a linear wave theory is not suitable to fully represent the behavior of a real sea. Therefore, the WEC must be designed in such a way that large system dynamics are to achieved in nonlinear random sea states. However, the current toolboxes for calculating fluid-structure interaction require a large computational effort to simulate interaction between floating bodies and fluids, and thus the motion of the floating bodies in nonlinear random waves. The aim of the proposed project is to create and validate an efficient simulation model that not only accounts for the mechanical part of the WEC, but also for the nonlinear random fluid-structure interaction between the floating body and the water. The created simulation model is used to analyze the dynamics of the WEC and the amount of energy gained by a WEC in nonlinear random sea states. The results are compared to the corresponding results in a linear random sea state. These comparisons are intended to demonstrate the necessity of using nonlinear random sea states. The used WEC is a mechanical system consisting of a floating cylindrical body and a generator, which has been examined in preliminary studies in linear sea state. The generator harvests energy from the motion of the cylinder. The behavior of the system is analyzed in simulations and validated in experiments. For experimental studies, the institute possesses a wave flume, in which random and nonlinear waves can be generated. In order to achieve high motion of the cylinder for a wide range of water waves, the simulation model is used to optimize several system parameters of the WEC (such as the radius and mass of the cylinder). In addition, control strategies are used to increase the amount of harvested energy. In this way, the dynamics of the WEC is optimized to extract a large amount of energy from random and nonlinear waves (as they occur in real sea states).
      28
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      18
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    Adaptive control of coupled rigid and flexible multibody systems with port-Hamiltonian structure
    The objective of the proposed research project is the development of adaptive tracking control techniques for coupled multibody systems with rigid and flexible components. The rigid components are described by nonlinear differential-algebraic equations. The flexible components are described by linear partial differential equations in one spatial dimension for a start. In the course of the project also multi-dimensional flexible subsystems will be considered, which are discretized by suitable methods. In each case, the models exhibit a port-Hamiltonian structure, and hence the physical properties (in particular, the power balance) can be captured in a mathematically rigorous way. A distinctive feature of port-Hamiltonian systems is that they are intrinsically modular, because arbitrary subsystems can be coupled via their ports. Despite these advantages, the port-Hamiltonian approach to modeling gets only little recognition in mechanics. Therefore, systematic methods for tracking control of such coupled multibody systems are still missing. In this project, first, a structural characterization of important system theoretic properties, such as input-output configurations, possible delays and the stability of the internal dynamics, will be conducted on the basis of physical considerations. Building on that, control techniques will be developed which guarantee the evolution of the tracking error within a prescribed margin. To this end, methods from funnel control and inversion-based feedforward control are combined, which led to far-reaching results for rigid multibody systems described by differential-algebraic equations in the first phase of the project. Now, in the second phase of the project, the feedforward controller will be designed for an approximation of the flexible components via coarse discretizations and combined with a funnel controller for the exact model. The latter is intended to compensate the approximation errors. Exploiting the modular construction of port-Hamiltonian systems, the possibility of a recursive feedforward control design will be investigated. For the applicability of funnel control, approaches which establish a functional relation between the original output and an alternative output that is co-located to the input will be considered. The performance and implementability of the developed methods will be constantly verified by means of selected experiments. The experiments support the selection of technically suitable controller design parameters and thus lead to a feedback between theory and practice.
      40
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      25
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
      28
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Project, Access status: without files ,
    DynManto - Modellierung und Simulation flexibler Mehrkörpersysteme
    DynManto (vormals SimFMBS) ist eine institutseigene Matlab-Toolbox für die Modellierung, Analyse und Optimierung von starren und flexiblen Mehrkörpersystemen. Die Grundidee dieses akademischen Softwareprojekts ist es, ein Simulationswerkzeug zu entwickeln, welches Studenten und Wissenschaftlern erlaubt, einfach und schnell Modelle für die Analyse von Mehrkörpersystemen zu erstellen und zu simulieren. Die Bewegungsdifferentialgleichungen werden dabei als differential-algebraische Gleichungen formuliert und mit Hilfe von Stabilisierung- oder Projektionsverfahren numerisch gelöst. Die Modellierung der flexiblen Körper erfolgt mit dem Ansatz des mitbewegten Referenzsystems.
      451