Anlehnend an die Entwicklung cyber-physischer Systeme können roboterbasierte Prüfsysteme durch zusätzliche Sensorik und die Integration kognitiver Systembausteine in ihrer Flexibilität und ihrem Automatisierungsgrad erhöht werden. In der industriellen Praxis ist dies für die Prüfung komplexer Produktmerkmale wie der haptischen Bedienqualität von besonderem Interesse. Entgegen der Prüfung konventioneller Produktmerkmale werden haptische Prüfungen gegenwärtig überwiegend manuell durchgeführt. Mit Blick auf die Prüfung wahrgenommener Qualität ergeben sich durch die Besonderheiten der menschlichen Wahrnehmung Anforderungen an die Prüfung, die über eine Objektivierung mittels messtechnischer Beschreibungsmerkmale hinausgehen. Psychophysikalische Konzepte der menschlichen Wahrnehmung wie eine einheitliche und aufeinander abgestimmte Bediencharakteristik von Bedienelementen innerhalb eines lokalen Bereiches müssen im gleichen Maße in der Auswertung berücksichtigt werden. Ausgehend von den Defiziten bestehender haptischer Prüfsysteme wird am Anwendungsbeispiel der Betätigungshaptik von Bedienelementen im KFZ-Innenraum die Entwicklung und Implementierung eines Prüfsystems vorgestellt, die zentrale Gestaltungsmerkmale cyber-physischer Systeme aufgreift.

Designing manufacturing systems requires a profound understanding of the manufacturing process and its challenges to meet final customer requirements. Considering future objectives already at an early design stage increases the flexibility of the manufacturing system and its robustness regarding changed boundary conditions. Today’s manufacturing systems rather control machine settings than process variables or even product quality. The major barrier for quality control is that in most manufacturing processes, quality cannot be measured on-line. Model-based self-sptimization (MBSO) has been developed to overcome this limitation. A combination of embedded process knowledge and tailored sensor integration enables for on-line quality estimation. The overall objective is to control key characteristics of product quality in a broad manufacturing landscape. This work describes a guideline of how to design an MBSO system with examples at each stage of the development process.

Purpose This paper aims to provide an approach of modeling haptic impressions of surfaces over a wide range of applications by using multiple sensor sources. Design/methodology/approach A multisensory measurement experiment was conducted using various leather and artificial leather surfaces. After processing of measurement data and feature extraction, different learning algorithms were applied to the measurement data and a corresponding set of data from a sensory study. The study contained evaluations of the same surfaces regarding descriptors of haptic quality (e.g. roughness) by human subjects and was conducted in a former research project. Findings The research revealed that it is possible to model and project haptic impressions by using multiple sensor sources in combination with data fusion. The presented method possesses the potential for an industrial application. Originality/value This paper provides a new approach to predict haptic impressions of surfaces by using multiple sensor sources.

Camera-based stereo-vision provides cost-efficient vision capabilities for robotic systems. The objective of this paper is to examine the performance of stereo-vision as means to enable a robotic inspection cell for haptic quality testing with the ability to detect relevant information related to the inspection task. This information comprises the location and 3D representation of a complex object under inspection as well as the location and type of quality features which are subject to the inspection task. Among the challenges is the low-distinctiveness of features in neighboring area, inconsistent lighting, similar colors as well as low intra-class variances impeding the retrieval of quality characteristics. The paper presents the general outline of the vision chain as well as performance analysis of various algorithms for relevant steps in the machine vision chain thus indicating the capabilities and drawbacks of a camera-based stereo-vision for flexible use in complex machine vision tasks.

No AccessHandbuch Industrie 4.0Apr 2017Organisation, Qualität und IT-Systeme für Planung und BetriebMichael Niehues, Gunther Reinhart, Robert H. Schmitt, Günther Schuh, Felix Brambring, Max Ellerich, Hannes Elser, Daniel Frank, Sebastian Groggert, Andreas Gützlaff, Verena Heinrichs, Thomas Hempel, Kevin Kostyszyn, Hao Ngo, Laura Niendorf, Eike Permin, Jan-Philipp Prote, Christina Reuter, Robin TürtmannMichael NiehuesSearch for more papers by this author, Gunther ReinhartSearch for more papers by this author, Robert H. SchmittSearch for more papers by this author, Günther SchuhSearch for more papers by this author, Felix BrambringSearch for more papers by this author, Max EllerichSearch for more papers by this author, Hannes ElserSearch for more papers by this author, Daniel FrankSearch for more papers by this author, Sebastian GroggertSearch for more papers by this author, Andreas GützlaffSearch for more papers by this author, Verena HeinrichsSearch for more papers by this author, Thomas HempelSearch for more papers by this author, Kevin KostyszynSearch for more papers by this author, Hao NgoSearch for more papers by this author, Laura NiendorfSearch for more papers by this author, Eike PerminSearch for more papers by this author, Jan-Philipp ProteSearch for more papers by this author, Christina ReuterSearch for more papers by this author, Robin TürtmannSearch for more papers by this authorhttps://doi.org/10.3139/9783446449893.006SectionsAboutPDF ToolsAdd to FavoritesDownload CitationTrack CitationsCopy LTI LinkPDF key 'share (en)' returned an object instead of string.FacebookTwitterEmailLinkedIn previous chapternext chapter FiguresReferencesRelatedDetails 2017Pages: 137-167Print ISBN: 978-3-446-44642-7eISBN: 978-3-446-44989-3 Copyright & Permissions© 2017 Carl Hanser Verlag GmbH & Co. KGPDF downloadLoading ...

Haptic perception is regarded as a key component of customer appreciation and acceptance for various products. The prediction of customers' haptic perception is of interest both during product development and production phases. This paper presents the results of a multivariate analysis between perceived roughness and texture related surface measurements, to examine whether perceived roughness can be accurately predicted using technical measurements. Studies have shown that standardized measurement parameters, such as the roughness coefficients (e.g. Rz or Ra), do not show a one-dimensional linear correlation with the human perception (of roughness). Thus, an alternative measurement method was compared to standard measurements of roughness, in regard to its capability of predicting perceived roughness through technical measurements. To estimate perceived roughness, an experimental study was conducted in which 102 subjects evaluated four sets of 12 different geometrical surface structures regarding their relative perceived roughness. The two different metrological procedures were examined in relation to their capability to predict the perceived roughness of the subjects stated within the study. The standardized measurements of the surface roughness were made using a structured light 3D-scanner. As an alternative method, surface induced vibrations were measured by a finger-like sensor during robot-controlled traverse over a surface. The presented findings provide a better understanding of the predictability of human haptic perception using technical measurements.

Managing the quality of a product and a company within a production network is a challenging task for decision makers, as the impact of actions is often time-delayed and can lead to the famous bullwhip effect in a multi stage supply chain. In order to qualify decision makers to understand the fundamental principles of quality management in production networks and thus enable them to effectively use it to optimize quality within his or her company, a game based simulation similar to Goldratt's game, called Q-I-Game, has been developed and tested in a previous research project. Within the game, one player acts as a factory between a supplier and a customer and states his desired invest into quality, the quantity of parts to be ordered and the investment into an inspection of incoming parts. The game uses simple, artificial functions to model effects between the investment into quality and the resulting quality of produced parts. To improve the learning effect of the game in regards to inspection planning as well as to promote understanding of the mechanics, the functionality of the game was expanded and a more realistic model of production is being developed and shown in this paper. The added functionality allows to simulate an extended supply chain with multiple players all acting as customer and supplier within the game. Thus, the game has evolved from simple single player to a complex multi-player game, taking the fact into consideration, that the bullwhip effect increases downstream of the supply chain. In addition, players will have the opportunity to make decisions on defect parts coming from a supplier (e.g. reject them and accept a new supply date). The extensions of the game which are already realised will be presented within the paper and future developments will be discussed.

Haptic perception of texture is an important tool in the evaluation of products such as smartphones, automobiles, kitchen appliances etc. Successful companies thus aim at actively designing the haptic impression of their products to the haptic preferences of their customers. However doing so requires information as to how comparable product surfaces are perceived by the customer and how these surfaces can be characterized technically. The perceived haptic impression of an object can be estimated using human studies, whereas its technical characteristics can be measured. However, direct correlation between perceived and standard measurements often shows poor significance, thus perceived haptic impression can not be simply derived from standard measurements. It remains to be analysed, if alternative measurements can provide a more significant characterization of surfaces in terms of their perceived haptic impression. To allow a more efficient estimation of the perceived impression of an object, an automatic inspection system is presented. This system consists of a force controlled robot driving a biomimetic sensor by the company Syntouch, providing a vibration signal during traverse over a surface. The paper presents results of a study to correlate the sensor signals to the perceived roughness of different surfaces evaluated by human subjects and set this in comparison to a correlation between standard roughness values and the perceived roughness.

The goal of this research was to determine the cause-and-effect relationships within a 3D printer and the applicability of experimental process models on generating optimum printing parameters, regardless of the printed 3D object. Four target quality parameters and seven factors were chosen to be examined experimentally. A set of hypotheses for cause-and-effect relationships was derived from the evaluation of the 3D-printing system, prior to the experiments being executed. A model was determined from the significant correlations to generate optimum sets of parameters. Five samples of two different 3D objects each were printed for two sets of optimization plots to validate the optimized parameter settings. The accuracy of the model predictions was evaluated in regard to the general applicability of the process model toward finding optimum process parameters. The results indicate that predictions from experimental process models of 3D printers remain mostly valid if they are used to predict target values for different 3D objects. A deviation of 7-9% was observed in the prediction of the surface quality. It can be concluded that a combination of the experimental model with existing expert knowledge and physical correlations in an advanced gray-box process model could enhance the accuracy and applicability of the experimental results.

In order to be competitive and to offer distinguishable products which delight customers, companies have to tailor their products to the exact needs and specifications of their customers. At the same time, companies have less time to develop new products or make changes to existing products, due to an overall shortening of product lifecycles and increasing market related cost pressure. Because of this lack of resources in product development, products are often not validated against the real customer impression until they reach the market. In consequence, companies risk nonacceptance of their products by the customer. Additional changes in product specifications might come too late and not make up for the lost trust of the customers in the company's products.To overcome this challenge, companies have to improve the product validation process, i.e. validate more quickly and earlier. A self-optimizing validation system could present a possible improvement in this regard. Using this efficient system, the validation could take place before the product is delivered to the market. In addition, the information from product validations would automatically be prepared and directed to the product development or the production process, to enable changes within the product specifications, without necessary interpretations of the validation. The paper gives answer to the question, how validation can be improved regarding efficiency and objectivity. To do so, the concept of a self-optimizing validation system using tactile sensors is presented, which can be used to validate the haptic product perception of customers. To display the relevant functions of this system and the necessary interfaces to the linked processes, a Viable Systems Model of the technical system is used to demonstrate this concept. The results implicate the applicability of the Viable Systems Model on technical systems and represent a contribution to the research towards a self-optimizing production system. (C) 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).