Kirchner, Frank

Preferred name

Kirchner, Frank

Official Name

Kirchner, Frank

Gerade angezeigt 1 - 10 von 18

Some of the metrics are blocked by your
consent settings
Generalizing, Decoding, and Optimizing Support Vector Machine Classification
(2015-03-26)
Krell, Mario Michael
;
Büskens, Christof
;
Kirchner, Frank
;
Kirchner, Frank
The classification of complex data usually requires the composition of processing steps. Here, a major challenge is the selection of optimal algorithms for preprocessing and classification. Nowadays, parts of the optimization process are automized but expert knowledge and manual work are still required. We present three steps to face this process and ease the optimization. Namely, we take a theoretical view on classical classifiers, provide an approach to interpret the classifier together with the preprocessing, and integrate both into one framework which enables a semiautomatic optimization of the processing chain and which interfaces numerous algorithms.
Dissertation
924 162
Some of the metrics are blocked by your
consent settings
Development, Control, and Empirical Evaluation of the Six-Legged Robot SpaceClimber Designed for Extraterrestrial Crater Exploration
(2013-02-20)
Bartsch, Sebastian
;
Hertzberg, Joachim
;
Kirchner, Frank
;
Kirchner, Frank
In the recent past, mobile robots played an important role in the field of extraterrestrial surface exploration. Unfortunately, the currently available space exploration rovers do not provide the necessary mobility to reach scientifically interesting places in rough and steep terrain like boulder fields and craters. Multi-legged robots have proven to be a good solution to provide high mobility in unstructured environments. However, space missions place high demands on the system design, control, and performance which are hard to fulfill with such kinematically complex systems. This thesis focuses on the development, control, and evaluation of a six-legged robot for the purpose of lunar crater exploration considering the requirements arising from the envisaged mission scenario. The performance of the developed system is evaluated and optimized based on empirical data acquired in significant and reproducible experiments performed in a laboratory environment in order to show thecapability of the system to perform such a task and to provide a basis for the comparability with other mobile robotic solutions.
Dissertation
434 440
Some of the metrics are blocked by your
consent settings
Autonomous Operation of a Reconfigurable Multi-Robot System for Planetary Space Missions
(2019-06-19)
Roehr, Thomas
;
Hertzberg, Joachim
;
Kirchner, Frank
;
Kirchner, Frank
Reconfigurable robots can physically merge and form new types of composite systems. This ability leads to additional degrees of freedom for robot operations especially when dynamically composed robotic systems offer capabilities that none of the individual systems have. Research in the area of reconfigurable multi-robot systems has mainly been focused on swarm-based robots and thereby to systems with a high degree of modularity but a heavily restricted set of capabilities. In contrast, this thesis deals with heterogeneous robot teams comprising individually capable robots which are also modular and reconfigurable. In particular, the autonomous application of such reconfigurable multi-robot systems to enhance robotic space exploration missions is investigated. Exploiting the flexibility of a reconfigurable multi-robot system requires an appropriate system model and reasoner. Hence, this thesis introduces a special organisation model. This model accounts for the key characteristics of reconfigurable robots which are constrained by the availability and compatibility of hardware interfaces. A newly introduced mapping function between resource structures and functional properties permits to characterise dynamically created agent compositions. Since a combinatorial challenge lies in the identification of feasible and functionally suitable agents, this thesis further suggests bounding strategies to reason efficiently with composite robotic systems. This thesis proposes a mission planning algorithm which permits to exploit the flexibility of reconfigurable multi-robot systems. The implemented planner builds upon the developed organisation model so that multi-robot missions can be specified by high-level functionality constraints which are resolved to suitable combinations of robots. Furthermore, the planner synchronises robot activities over time and characterises plans according to three objectives: efficacy, efficiency and safety. The plannera s evaluation demonstrates an optimization of an exemplary space mission. This research is based on the parallel development of theoretical concepts and practical solutions while working with three reconfigurable multi-robot teams. The operation of a reconfigurable robotic team comes with practical constraints. Therefore, this thesis composes and evaluates an operational infrastructure which can serve as reference implementation. The identification and combination of applicable state-of-the-art technologies result in a distributed and dynamically extensible communication infrastructure which can maintain the properties of reconfigurable multi-robot systems. Field tests covering semi-autonomous and autonomous operation have been performed to characterise multi-robot missions and validate the autonomous control approach for reconfigurable multi-robot systems. The practical evaluation identified critical constraints and design elements for a successful application of reconfigurable multi-robot systems. Furthermore, the experiments point to improvements for the organisation model. This thesis is a wholistic approach to automate reconfigurable multi-robot systems. It identifies theoretical as well as practical challenges and it suggests effective solutions which permit an exploitation of an increased level of flexibility in future robotics missions.
Dissertation
1624 698
Some of the metrics are blocked by your
consent settings
Design and development of a hominid robot with local control in its adaptable feet to enhance locomotion capabilities
(2016-12-05)
Kühn, Daniel
;
Frese, Udo
;
Kirchner, Frank
;
Kirchner, Frank
With increasing mechanization of our daily lives, the expectations and demands in robotic systems increase in the general public and in scientists alike. In recent events such as the Deepwater Horizon''-accident or the nuclear disaster at Fukushima, mobile robotic systems were used, e.g., to support local task forces by gaining visual material to allow an analysis of the situation. Especially the Fukushima example shows that the robotic systems not only have to face a variety of different tasks during operation but also have to deal with different demands regarding the robot's mobility characteristics. To be able to cope with future requirements, it seems necessary to develop kinematically complex systems that feature several different operating modes. That is where this thesis comes in: A robotic system is developed, whose morphology is oriented on chimpanzees and which has the possibility due to its electro-mechanical structure and the degrees of freedom in its arms and legs to walk with different gaits in different postures. For the proposed robot, the chimpanzee was chosen as a model, since these animals show a multitude of different gaits in nature. A quadrupedal gait like crawl allows the robot to traverse safely and stable over rough terrain. A change into the humanoid, bipedal posture enables the robot to move in man-made environments. The structures, which are necessary to ensure an effective and stable locomotion in these two poses, e.g., the feet, are presented in more detail within the thesis. This includes the biological model and an abstraction to allow a technical implementation. In addition, biological spines are analyzed and the development of an active, artificial spine for the robotic system is described. These additional degrees of freedom can increase the robot's locomotion and manipulation capabilities and even allow to show movements, which are not possible without a spine. Unfortunately, the benefits of using an artificial spine in robotic systems are nowadays still neglected, due to the increased complexity of system design and control. To be able to control such a kinematically complex system, a multitude of sensors is installed within the robot's structures. By placing evaluation electronics close by, a local and decentralized preprocessing is realized. Due to this preprocessing is it possible to realize behaviors on the lowest level of robot control: in this thesis it is exemplarily demonstrated by a local controller in the robot's lower leg. In addition to the development and evaluation of robot's structures, the functionality of the overall system is analyzed in different environments. This includes the presentation of detailed data to show the advantages and disadvantages of the local controller. The robot can change its posture independently from a quadrupedal into a bipedal stance and the other way around without external assistance. Once the robot stands upright, it is to investigate to what extent the quadrupedal walking pattern and control structures (like the local controller) have to be modified to contribute to the bipedal walking as well.
Dissertation
672 358
Some of the metrics are blocked by your
consent settings
Modular and Analytical Methods for Solving Kinematics and Dynamics of Series-Parallel Hybrid Robots
(2019-11-15)
Kumar, Shivesh
;
Mueller, Andreas
;
Kirchner, Frank
;
Kirchner, Frank
While serial robots are known for their versatility in applications, larger workspace, simpler modeling and control, they have certain disadvantages like limited precision, lower stiffness and poor dynamic characteristics in general. A parallel robot can offer higher stiffness, speed, accuracy and payload capacity, at the downside of a reduced workspace and a more complex geometry that needs careful analysis and control. To bring the best of the two worlds, parallel submechanism modules can be connected in series to achieve a series-parallel hybrid robot with better dynamic characteristics and larger workspace. Such a design philosophy is being used in several robots not only at DFKI (for e.g., Mantis, Charlie, Recupera Exoskeleton, RH5 humanoid etc.) but also around the world, for e.g. Lola (TUM), Valkyrie (NASA), THOR (Virginia Tech.) etc.These robots inherit the complexity of both serial and parallel architectures. Hence, solving their kinematics and dynamics is challenging because they are subjected to additional geometric loop closure constraints. Most approaches in multi-body dynamics adopt numerical resolution of these constraints for the sake of generality but may suffer from inaccuracy and performance issues. They also do not exploit the modularity in robot design. Further, closed loop systems can have variable mobility, different assembly modes and can impose redundant constraints on the equations of motion which deteriorates the quality of many multi-body dynamics solvers. Very often only a local view to the system behavior is possible. Hence, it is interesting for geometers or kinematics researchers, to study the analytical solutions to geometric problems associated with a specific type of parallel mechanism and their importance over numerical solutions is irrefutable. Techniques such as screw theory, computational algebraic geometry, elimination and continuation methods are popular in this domain. But this domain specific knowledge is often underrepresented in the design of model based kinematics and dynamics software frameworks. The contributions of this thesis are two-fold. Firstly, a rigorous and comprehensive kinematic analysis is performed for the novel parallel mechanisms invented recently at DFKI-RIC such as RH5 ankle mechanism and Active Ankle using approaches from computational algebraic geometry and screw theory. Secondly, the general idea of a modular software framework called Hybrid Robot Dynamics (HyRoDyn) is presented which can be used to solve the geometry, kinematics and dynamics of series-parallel hybrid robotic systems with the help of a software database which stores the analytical solutions for parallel submechanism modules in a configurable and unit testable manner. HyRoDyn approach is suitable for both high fidelity simulations and real-time control of complex series-parallel hybrid robots. The results from this thesis has been applied to two robotic systems namely Recupera-Reha exoskeleton and RH5 humanoid. The aim of this software tool is to assist both designers and control engineers in developing complex robotic systems of the future. Efficient kinematic and dynamic modeling can lead to more compliant behavior, better whole body control, walking and manipulating capabilities etc. which are highly desired in the present day and future robotic applications.
Dissertation
1008 1195
Some of the metrics are blocked by your
consent settings
Object Recognition and Localization : the Role of Tactile Sensors
(2015-02-25)
Aggarwal, Achint
;
Frese, Udo
;
Kirchner, Frank
;
Kirchner, Frank
Tactile sensors, because of their intrinsic insensitivity to lighting conditions and water turbidity, provide promising opportunities for augmenting the capabilities of vision sensors in applications involving object recognition and localization. This thesis presents two approaches for haptic object recognition and localization for ground and underwater environments. The first approach called Batch Ransac and Iterative Closest Point augmented Sequential Filter (BRICPSF) is based on an innovative combination of a sequential filter, Iterative-Closest-Point algorithm, and a feature-based Random Sampling and Consensus (RANSAC) algorithm for database matching. It can handle a large database of 3D-objects of complex shapes and performs a complete six-degree-of-freedom localization of static objects. The algorithms are validated by experimentation in simulation and using actual hardware. To our knowledge this is the first instance of haptic object recognition and localization in underwater environments. The second approach is biologically inspired, and provides a close integration between exploration and recognition. An edge following exploration strategy is developed that receives feedback from the current state of recognition. A recognition by parts approach is developed which uses BRICPSF for object part recognition. Object exploration is either directed to explore a part until it is successfully recognized, or is directed towards new parts to endorse the current recognition belief. This approach is validated by simulation experiments.
Dissertation
487 283
Some of the metrics are blocked by your
consent settings
Learning the Structure of Continuous Markov Decision Processes
(2014-02-21)
Metzen, Jan Hendrik
;
Kreowski, Hans-Jörg
;
Kirchner, Frank
;
Kirchner, Frank
There is growing interest in artificial, intelligent agents which can operate autonomously for an extended period of time in complex environments and fulfill a variety of different tasks. Such agents will face different problems during their lifetime which may not be foreseeable at the time of their deployment. Thus, the capacity for lifelong learning of new behaviors is an essential prerequisite for this kind of agents as it enables them to deal with unforeseen situations. However, learning every complex behavior anew from scratch would be cumbersome for the agent. It is more plausible to consider behavior to be modular and let the agent acquire a set of reusable building blocks for behavior, the so-called skills. These skills might, once acquired, facilitate fast learning and adaptation of behavior to new situations. This work focuses on computational approaches for skill acquisition, namely which kind of skills shall be acquired and how to acquire them. The former is commonly denoted as "skill discovery" and the latter as "skill learning". The main contribution of this thesis is a novel incremental skill acquisition approach which is suited for lifelong learning. In this approach, the agent learns incrementally a graph-based representation of a domain and exploits certain properties of this graph such as its bottlenecks for skill discovery. This thesis proposes a novel approach for learning a graph-based representation of continuous domains based on formalizing the problem as a probabilistic generative model. Furthermore, a new incremental agglomerative clustering approach for identifying bottlenecks of such graphs is presented. Thereupon, the thesis proposes a novel intrinsic motivation system which enables an agent to intelligently allocate time between skill discovery and skill learning in developmental settings, where the agent is not constrained by external tasks. The results of this thesis show that the resulting skill acquisition approach is suited for continuous domains and can deal with domain stochasticity and different explorative behavior of the agent. The acquired skills are reusable and versatile and can be used in multi-task and lifelong learning settings in high-dimensional problems.
Dissertation
629 188
Some of the metrics are blocked by your
consent settings
An Intelligent Architecture for Legged Robot Terrain Classification Using Proprioceptive and Exteroceptive Data
(2015-06-24)
Ahmed, Mohammed Nour Abdelgwad
;
Frese, Udo
;
Kirchner, Frank
;
Kirchner, Frank
In this thesis, we introduce a novel architecture called "Intelligent Architecture for Legged Robot Terrain Classification Using Proprioceptive and Exteroceptive Data (iARTEC )". The proposed architecture integrates different terrain characterization and classification with other robotic system components. Within iARTEC , we consider the problem of having a legged robot autonomously learn to identify different terrains. Robust terrain identification can be used to enhance the capabilities of legged robot systems, both in terms of locomotion and navigation. For example, a robot that has learned to differentiate sand from gravel can autonomously modify (or even select a different) path in favor of traversing over a better terrain. The same knowledge of the terrain type can also be used to guide a robot in order to avoid specific terrains. To tackle this problem, we developed four approaches for terrain characterization, classification, path planning, and control for a mobile legged robot. We developed a particle system inspired approach to estimate the robot foot ground contact interaction forces. The approach is derived from the well known Bekker s theory to estimate the contact forces based on its point contact model concepts. It is realistically model real-time 3-dimensional contact behaviors between rigid body objects and the soil. For a real-time capable implementation of this approach, its reformulated to use a lookup table generated from simple contact experiments of the robot foot with the terrain. Also, we introduced a short-range terrain classifier using the robot embodied data. The classifier is based on a supervised machine learning approach to optimize the classifier parameters and terrain it using proprioceptive sensor measurements. The learning framework preprocesses sensor data through channel reduction and filtering such that the classifier is trained on the feature vectors that are closely associated with terrain class. For the long-range terrain type prediction using the robot exteroceptive data, we present an online visual terrain classification system. It uses only a monocular camera with a feature-based terrain classification algorithm which is robust to changes in illumination and view points. For this algorithm, we extract local features of terrains using Speed Up Robust Feature (SURF). We encode the features using the Bag of Words (BoW) technique, and then classify the words using Support Vector Machines (SVMs). In addition, we described a terrain dependent navigation and path planning approach that is based on E* planer and employs a proposed metric that specifies the navigation costs associated terrain types. This generated path naturally avoids obstacles and favors terrains with lower values of the metric. At the low level, a proportional input-scaling controller is designed and implemented to autonomously steer the robot to follow the desired path in a stable manner. iARTEC performance was tested and validated experimentally using several different sensing modalities (proprioceptive and exteroceptive) and on the six legged robotic platform CREX. The results show that the proposed architecture integrating the aforementioned approaches with the robotic system allowed the robot to learn both robot-terrain interaction and remote terrain perception models, as well as the relations linking those models. This learning mechanism is performed according to the robot own embodied data. Based on the knowledge available, the approach makes use of the detected remote terrain classes to predict the most probable navigation behavior. With the assigned metric, the performance of the robot on a given terrain is predicted. This allows the navigation of the robot to be influenced by the learned models. Finally, we believe that iARTEC and the methods proposed in this thesis can likely also be implemented on other robot types (such as wheeled robots), although we did not test this option in our work.
Dissertation
319 192
Some of the metrics are blocked by your
consent settings
Constraint basierte Modellierung von Komponentennetzwerken am Beispiel eines autonomen Unterwasserfahrzeugs
(2017-08-17)
Goldhoorn, Matthias
;
Frese, Udo
;
Kirchner, Frank
;
Kirchner, Frank
Moderne Robotersysteme werden zunehmend komplexer. Die Forschung im Bereich der Robotik sieht sich immer mehr mit dem Problem der Integration von Algorithmen in neue Plattformen konfrontiert. Es gibt zahlreiche innovative Verfahren und Algorithmen, die die Robotersysteme mit neuen Fähigkeiten ausstatten. Diese Erweiterungen lassen sich sowohl auf der Hardware- als auch der Softwareebene erkennen. Jedoch findet diese Vielzahl an neuen Entwicklungen nur selten Anwendung auf den aktuellen Ziel-Plattformen. Dies bremst weitere Entwicklungen, da mehr Zeit für die Integration als für die eigentliche Forschung aufgewendet wird. Mit einem Blick über die eigene Forschung hinaus oder durch die detailliertere Beschäftigung mit den eingesetzten Verfahren lässt sich erkennen, dass die Lösungen zumeist systemspezifisch sind oder sich nur mit großer Mühe auf andere oder gar größere komplexere Maßstäbe übertragen lassen. Das Ziel dieser Arbeit ist es daher zu identifizieren, wo die größten Herausforderun liegen. Im Anschluss soll das markanteste Problem gelöst werden. Die Modularisierung sowie die modellgetriebene Entwicklung sind dabei die größten Änderungen, die sich Abzeichen. Beide Trends haben verschiedene, teils gravierende Einflüsse auf die Art und Weise, wie Entwicklung stattfindet. Modularisierung erhöht die Wiederverwendbarkeit einzelner Module und führt zu kleineren Verbunden. Modellgetriebene Entwicklung versucht mögliche Fehler in Konzepten zu einem frühen Zeitpunkt zu erkennen. Konkret bedeuten beide Verfahren jedoch Mehrarbeit im Vergleich zu der klassischen monolithischen und reaktiven Systementwicklung. Die Komponenten lassen sich zwar oftmals wiederverwenden, aber die Konsistenz und Nebeneffekte der Selektion bzw. der Abhängigkeiten erfordern viel Zeit und Expertenwissen. Die modellgetriebene Softwareentwicklung verlangt von den Entwicklern von Algorithmen bereits im Voraus Informationen darüber, wie und unter welchen Bedingungen ihr Algorithmus operieren kann. Auf der anderen Seite, der Hardwareentwicklung, stehen immer kleinere und hoch integrierte Sensoren und Aktuatoren zur Verfügung, die wiederum die Konstruktion immer komplexerer Systeme erlauben. Die Systeme bieten oftmals ausreichend Redundanz, um Probleme auf unterschiedliche Art und Weise lösen zu können.
Dissertation
349 223
Some of the metrics are blocked by your
consent settings
Design and Experimental Evaluation of a Hybrid Wheeled-Leg Exploration Rover in the Context of Multi-Robot Systems
(2018-12-12)
Cordes, Florian
;
Frese, Udo
;
Kirchner, Frank
;
Kirchner, Frank
With this dissertation, the electromechanic design, implementation, locomotion control, and experimental evaluation of a novel type of hybrid wheeled-leg exploration rover are presented. The actively articulated suspension system of the rover is the basis for advanced locomotive capabilities of a mobile exploration robot. The developed locomotion control system abstracts the complex kinematics of the suspension system and provides platform control inputs usable by autonomous behaviors or human remote control. Design and control of the suspension system as well as experimentation with the resulting rover are in the focus of this thesis. The rover is part of a heterogeneous modular multi-robot exploration system with an aspired sample return mission to the lunar south pole or currently hard-to-access regions on Mars. The multi-robot system pursues a modular and reconfigurable design methodology. It combines heterogeneous robots with different locomotion capabilities for enhanced overall performance. Consequently, the design of the multi-robot system is presented as the frame of the rover developments. The requirements for the rover design originating from the deployment in a modular multi-robot system are accentuated and summarized in this thesis.
Dissertation
747 275

Kirchner, Frank

Filter

Einstellungen

Sortieren nach

Ergebnisse pro Seite