Refine
Year of publication
Document Type
- Conference Proceeding (40)
- Article (14)
- Part of a Book (5)
- Other (3)
- Book (1)
- Doctoral Thesis (1)
Keywords
- arbeitsplatz (1)
- mensch-roboter (1)
- montage (1)
- robotik (1)
Institute
Im Zuge der fortschreitenden Globalisierung und zunehmenden Digitalisierung der Arbeitswelt, ergeben sich gerade für kleine und mittelständische Unternehmen Herausforderungen im Bereich der Produktion und insbesondere in der Montage. Steigende Komplexität der Produkte, kürzer werdende Produktlebenszyklen bei kleinen Losgrößen mit hoher Variantenvielfalt und großem Wettbewerbsdruck zwingen Unternehmen bereits vorhandene Montagestrategien zu überarbeiten. Gerade bei komplexeren Baugruppen ist eine Hochautomatisierung der Montage in der Serienfertigung aufgrund der Produktstruktur nur schwer realisierbar und oftmals nicht wirtschaftlich. In der industriellen Produktion ist ein klarer Trend von der Massenproduktion hin zur »Massenspezialanfertigung« zu erkennen. Die Leistungsfähigkeit eines Industriebetriebes hängt entscheidend von den angewandten Produktionsverfahren, den eingesetzten Produktionsmitteln und der eingeführten Produktionsorganisation ab. Aber auch das Zusammenspiel von Mensch, Organisation und Technik trägt dazu bei Potenziale für den Erfolg eines Unternehmens auszuschöpfen. Aufgrund sich ständig ändernder Marktbedingungen und der kostengünstigen Herstellung von Produkten in Niedriglohnländern bei gleichzeitig steigendem Qualitätsniveau müssen Produktionsstrukturen ständig neu überdacht und weiterentwickelt werden. Die Montage ist im Vergleich zur klassischen Fertigung (wie z.B. Drehen, Fräsen oder Schleifen) ein eher niedrig automatisierter Bereich der industriellen Produktion, indem allerdings eine hohe Wertschöpfung am Produkt stattfindet. Die manuelle Montage ist daher auch zukünftig ein wichtiger Bestandteil der industriellen Produktion. Um am Markt bestehen zu können, müssen Unternehmen nach Möglichkeiten suchen, um manuelle Arbeit produktiver und damit kosteneffizienter zu gestalten ohne dabei Abstriche bei der Qualität hinzunehmen. Zur Unterstützung manueller Tätigkeiten sind in den vergangenen Jahren verschiedene Innovationen in den Fokus der Unternehmen gerückt. Das Ziel muss es sein, den (Montage-)Standort Deutschland durch innovative Konzepte wie kollaborative Mensch-Roboter-Arbeitsplätze zu sichern.
Das IDEE ist mit vier Forschungsprofessoren besetzt, die durch eine Lehrentlastung Freiräume für den Auf- und Ausbau von interdisziplinären Forschungsaktivitäten nutzen können. Das Center Robotik (CERI) ist eines der Aufgabengebiete des IDEEs mit den beiden Schwerpunkten »Robotergestützte Arbeitsplätze der Zukunft« sowie »Intralogistik in der Smart Factory«.
Dieser Artikel beschäftigt sich mit dem zweiten Schwerpunkt. Anhand der CERI Aktivitäten im Bereich der fahrerlosen Transportsysteme wird aufgezeigt, wie der »Dreiklang« Forschung, Lehre und Transfer in der Praxis umgesetzt wird und die drei Bestandteile sich gegenseitig befruchten können.
Highly autonomous production cells are a crucial part of manufacturing systems in industry 4.0 and can contribute to a sustainable value-adding process. To realize a high degree of autonomy in production cells with an industrial robot and a machine tool, an experimental approach was carried out to deal with numerous challenges on various automation levels. One crucial aspect is the scheduling problem of tasks for each resource (machine tool, tools, robot, AGV) depending on various data needed for a job-shop scheduling algorithm. The findings show that the necessary data has to be derived from different automation levels in a company: horizontally from ERP to shop-floor, vertically from the order handling department to the maintenance department. Utilizing that data, the contribution provides a cascaded scheduling approach for machine tool jobs as well as CNC and robot tasks for highly autonomous production cells supplied by AGVs.
Decentralised estimation of heterogeneous sensors is performed on an outdoor network. Attributes such as position, appearance, and identity represented by non-Gaussian distributions are used in in the fusion process. It is shown here that real-time decentralised data fusion of non-Gaussian estimates can be used to build rich environmental maps. Human operators are also used as additional sensors in the network to complement robotic information.
This thesis is concerned with combining the perceptual abilities of mobile robots and human operators to execute tasks cooperatively. It is generally agreed that a synergy of human and robotic skills offers an opportunity to enhance the capabilities of today’s robotic systems, while also increasing their robustness and reliability. Systems which incorporate both human and robotic information sources have the potential to build complex world models, essential for both automated and human decision making. In this work, humans and robots are regarded as equal team members who interact and communicate on a peer-to-peer basis. Human-robot communication is addressed using probabilistic representations common in robotics. While communication can in general be bidirectional, this work focuses primarily on human-to-robot information flow. More specifically, the approach advocated in this thesis is to let robots fuse their sensor observations with observations obtained from human operators. While robotic perception is well-suited for lower level world descriptions such as geometric properties, humans are able to contribute perceptual information on higher abstraction levels. Human input is translated into the machine representation via Human Sensor Models. A common mathematical framework for humans and robots reinforces the notion of true peer-to-peer interaction. Human-robot information fusion is demonstrated in two application domains: (1) scalable information gathering, and (2) cooperative decision making. Scalable information gathering is experimentally demonstrated on a system comprised of a ground vehicle, an unmanned air vehicle, and two human operators in a natural environment. Information from humans and robots was fused in a fully decentralised manner to build a shared environment representation on multiple abstraction levels. Results are presented in the form of information exchange patterns, qualitatively demonstrating the benefits of human-robot information fusion. The second application domain adds decision making to the human-robot task. Rational decisions are made based on the robots’ current beliefs which are generated by fusing human and robotic observations. Since humans are considered a valuable resource in this context, operators are only queried for input when the expected benefit of an observation exceeds the cost of obtaining it. The system can be seen as adjusting its autonomy at run-time based on the uncertainty in the robots’ beliefs. A navigation task is used to demonstrate the adjustable autonomy system experimentally. Results from two experiments are reported: a quantitative evaluation of human-robot team effectiveness, and a user study to compare the system to classical teleoperation. Results show the superiority of the system with respect to performance, operator workload, and usability.
This paper describes an indoor Active Sensor Network, focussing on the implementation aspects of the system, including communication and the application framework. To make the system description more tangible we describe the latest in a series of indoor experiments implemented using ASN. The task is to detect and map motion of people (and robots) in an office space using a network of 12 stationary sensors. The network was operational for several days, with individual platform coming on and off line. On several occasions the network consisted of 39 components. The paper includes a section on the lessons learned during the project’s design and development which may be applicable to other heterogeneous distributed systems with data-intensive algorithms.
This Chapter describes Orca: an open-source project which applies Component-Based Software Engineering principles to robotics. It provides the means for defining and implementing interfaces such that components developed independently are likely to be inter-operable. In addition it provides a repository of free re-useable components. Orca attempts to be widely applicable by imposing minimal design constraints. This Chapter describes lessons learned while using Orca and steps taken to improve the framework based on those lessons. Improvements revolve around middleware issues and the problems encountered while scaling to larger distributed systems. Results are presented from systems that were implemented.
To arrive at a realistic assessment of localization methods in terms of their performance in an industrial environment under various challenging conditions, we provide a benchmark to evaluate algorithms both for individual components as well as multi-sensor systems. For several sensor types, including wheel odometry, RGB cameras, RGB-D cameras, and LIDAR, potential issues were identified. The accuracy of wheel odometry, for example, when there are bumps on the track. For each sensor type, we explicitly chose a track for the benchmark dataset containing situations where the sensor fails to provide adequate measurements. Based on the acquired sensor data, localization can be achieved either using a single sensor information or sensor fusion. To help evaluate the output of associated localization algorithms, we provide a software to evaluate a set of metrics as part of the paper. An example application of the benchmark with state-of-the-art algorithms for each sensor is also provided.
In this paper, an analysis of the precision of LIDAR range measurements is presented. LIDAR data from two different sensors (HLS-LFCD-LDS and SICK TIM561) were analyzed regarding the influence of range, incident angle to the surface, and material. Based on the results, a data-driven model for LIDAR precision behavior was developed, and a comparison with standard deviation models based on the vendor-provided specifications was presented. Our model can be used to create realistic sensor simulations and to develop robot navigation algorithms weighing sensor range readings based on the precision.
In this paper, we present the application of a non-linear dimensionality reduction technique for the learning and probabilistic classification of hyperspectral image. Hyperspectral image spectroscopy is an emerging technique for geological investigations from airborne or orbital sensors. It gives much greater information content per pixel on the image than a normal colour image. This should greatly help with the autonomous identification of natural and manmade objects in unfamiliar terrains for robotic vehicles. However, the large information content of such data makes interpretation of hyperspectral images time-consuming and userintensive. We propose the use of Isomap, a non-linear manifold learning technique combined with Expectation Maximisation in graphical probabilistic models for learning and classification. Isomap is used to find the underlying manifold of the training data. This low dimensional representation of the hyperspectral data facilitates the learning of a Gaussian Mixture Model representation, whose joint probability distributions can be calculated offline. The learnt model is then applied to the hyperspectral image at runtime and data classification can be performed.
This paper presents an approach of integrating human operators into a sensor network formed by a heterogeneous team of unmanned air and ground vehicles. Several objectives of human-network interaction are identified. The main focus of this work is on human-to-network information flow, i.e. human operators are regarded as information sources. It is argued that operators should make raw observations which are converted into the sensor network's common representation by a probabilistic model. The concepts are discussed in the context of an outdoor sensor network under development. Human operators contribute geometric feature information in the form of range and bearing observations. Visual feature properties are specified via meaningful class labels. A sensor model, represented as a Bayesian network, translates label observations into the system's representation. The model is also used to classify features as observed by robotic sensors.
This paper proposes a methodology to measure the effectiveness of a human-robot team as part of an adjustable autonomy system. The effectiveness measure is aimed at determining an appropriate autonomy level prior to the system's deployment. Two competing goals need to be traded off: maximising robot performance while minimising the amount of human input. The relative importance of the two goals depend on the mission priorities and constraints which are taken into account. The proposed methodology is applied to a human-robot communication system developed for task- oriented information exchange. The robot uses a decision- theoretic framework to act autonomously and to decide when to request input from human operators. The latter is achieved by computing the value-of-information an operator is able to provide which is compared to the cost of obtaining the information. For our system, the cost parameter represents the autonomy level to be determined. We demonstrate how an appropriate autonomy level can be found experimentally using a navigation task. In our experiment, the robot navigates through a set of simulated worlds with human input being generated by a software component. The results are used to find appropriate autonomy levels for three example missions and a subsequent user study.
This paper considers the modes of interaction between one or several human operators and an active sensor network-a fully decentralized network of sensors some or all of which have actuators and are in that sense active. The primary goal of this study is to investigate the conditions under which the human involvement will not jeopardize scalability of the overall system. Two aspects of human-robot interaction are considered: the ways in which the global view of the system may be conveyed to the operators, and how the operators may influence the behavior of the system during the course of its operation. The results of analysis favor peer-to-peer environment-centric interactions between the operators and the network whereby the humans act as extended sensors and communication nodes of the network itself. Experiments on an indoor active sensor network are described.
Humans and robots need to exchange information if the objective is to achieve a task cooperatively. Two questions are considered in this paper: what type of information to communicate, and how to cope with the limited resources of human operators. Decision-theoretic human-robot communication can provide answers to both questions: the type of information is determined by the underlying probabilistic representation, and value-of-information theory helps decide when it is appropriate to query operators for information. A robot navigation task is used to evaluate the system by comparing it to conventional teleoperation. The results of a user study show that the developed system is superior with respect to performance, operator workload, and usability.
This paper presents a novel people detection approach for mobile robot applications based on a combination of classical computer vision techniques and a state-of-the-art neural network. Our approach involves an RGB-D camera as an environmental sensor. The depth data is used to extract silhouettes around people. The RGB images are subsequently augmented with this border information before passing it to the neural network. Under challenging lighting conditions, our system was able to outperform the neural network trained on regular RGB data alone by a factor of two.
Today's reusable robotics software is provided by several self-contained open-source projects with virtually no software reuse between them. Such partitioning leads to problems with software quantity, quality, ease of evaluation and, ultimately, to poor end user experience. By reviewing several of the projects we observe that all of them contain a mix of three types of software: 1) driver and algorithm implementations, 2) communication middleware, and 3) robotic software framework. We show that more than half of the combined code base contains software which could be highly reusable but only a small fraction of it actually is. We argue that formal separation of the three groups in the existing and future software projects would offer several potential advantages. Availability of framework-independent code would enable community-wide library-based software reuse in addition to the existing framework-wide component- based reuse. Another important benefit is related to evaluation procedures. The three software types are very different and should be evaluated separately, using different criteria. The first two types allow quantitative comparisons which are well documented in the literature. The last one is largely qualitative and, therefore, more subjective. Practically, we recommend today's projects to refactor and eject the driver and algorithm implementation code and focus on the task of software integration. Thin frameworks can be numerous, increasing the diversity of options available to robotics practitioners. Finally, we discuss distribution options for the refactored framework-independent code including the option of creating a Common Robotic Project.