Search

This paper presents the results of an analysis of systematic development approaches for cognitive products found in literature and extracted from interviews with 13 researchers in academic research projects. Based on a harmonized order of applied development steps in the academic research projects and the understanding and expertise of their developers, a procedural model supporting the development of cognitive products is proposed, emphasizing the specific development steps required for cognitive products. The proposed procedural model builds on and is structured according to the V-model.

Diese Veröffentlichung erweitert die Entwicklung kognitiver Produkte um eine Bewertungsmethodik zur systematischen Integration von kognitiven Funktionen. Bisher existiert ein Vorgehen, das potentiell integrierbare kognitive Funktionen anhand der Nutzerinteraktion aufdeckt und systematisch darstellt. Die Untersuchung und Auswahl, welche Funktion in das Produktkonzept aufgenommen werden sollte, wird allerdings nicht ausreichend unterstützt und basiert bisher allein auf der Erfahrung der Entwickler. Deshalb ist, basierend auf etablierten Analyse- und Bewertungsmethoden in der Produktentwicklung, eine Bewertungsmethodik entwickelt worden, die es erlaubt, Eigenschaften von potentiell integrierbaren kognitiven Funktionen hinsichtlich vorab festgelegter Kriterien zu bewerten und zu priorisieren. Dabei werden sowohl die Bedürfnisse des Kunden als auch das unternehmensspezifische Entwicklungs-Know-how berücksichtigt. Das Ergebnis wird in einem Ergebnisportfolio dargestellt und ermöglicht systematisch und nachvollziehbar die Integration der erfolgversprechendsten kognitiven Funktionen.

Cognitive products are tangible and durable things with cognitive capabilities that meet and exceed user expectations by using cognitive functions, e.g. to perceive, to learn, to reason, etc., to reduce the need for human input. This paper presents a model-based assessment of the solution space for cognitive products. So far, the design of cognitive products has been based on prototype-oriented approaches, which mainly focus on cognitive algorithms, relying too much on designer´s experience, beliefs or ad hoc arbitrated processes and following as a consequence the “design it now and fix it later!”–philosophy. A model-based assessment of the solution space would enable a better and early estimation of design alternatives that meet not only software requirements but also hardware requirements from the very early stages down to system structural and behavioural aspects in highly dynamic and uncertain environments. The conventional MBSE approach has been adapted to cognitive products and is demonstrated using a cognitive coffee robot waiter.

Cognitive products use cognitive functions to work autonomously and reduce the amount of interaction necessary from the user. However, to date no method exists to support the integration of cognitive functions in common products. This paper presents a method that supports designers when exploring ideas for new cognitive products. The method is based on functions/actions that humans perform while using a product, as well as functions/actions performed by the product itself, all of which can be consistently modelled in an activity diagram. Initially, the system boundary of the product is drawn around the functions/actions performed by the product. Cognitive functions are then identified that are currently performed by the user, and can possibly be integrated into a new cognitive concept. The resulting concept is specified systematically by interpreting the system boundary of the product to include cognitive functions. This method has been verified via design projects performed by interdisciplinary student desig teams, and an example of this work is presented.

This paper presents a methodology to identify meaningful cognitive func-tions for existing products and thus turn existing product concepts into cog-nitive product concepts. The approach also works well to develop companion robots, e. g. for elderly care. One challenge when applying the methodology is to identify and integrate the right cognitive functions into the new cognitive product. The goal is to create an easy and comprehensible mental model that makes human-robot interaction worthwhile and relieve the users from repetitive, dangerous and difficult tasks by using cognitive functions without creating paternalism. Designers that use the mentioned methodology must be aware that the integration of every cognitive function can change the human-robot interaction dramatically. The decision about which cognitive functions are integrated is crucial for the acceptance of the resulting product. This is particularly true for the development of companion robots that closely interact with humans. Amongst others, the approach has been applied to develop two companion robots that increase the freedom of movement of elderly people without stigmatizing them and adapting to the user specific cognitive and physical abilities. This paper describes briefly the methodology to integrate cognitive functions into cognitive product concepts before presenting an application example, two prototypes and discussing the need to consider synchronized and reciprocal human-robot interaction already during the conceptual design phase of a companion robot.

This paper introduces Gertie the Robotic Desk Lamp, a novel research platform that has five degrees of freedom, and is equipped with a camera and microphone in its lamp shade. These features mean that Gertie is a flexible and low-cost resource for conducting research into cognitive products and human-robot interaction. It will be available as an open hardware on http://www.opengertie.org/. Gertie was designed from first principles, and assembled using off the shelf electronic components and parts fabricated using a 3D printer. In this paper, the design of Gertie is presented, and its application as a research platform is described. Gertie has already been used to investigate a problem of simple object tracking, building on computer vision algorithms. Furthermore, it has also been used to investigate and replicate emotional body language. By imitating human body language Gertie is capable of expressing four of the basic Ekman emotions: 1) joy; 2) sadness; 3) surprise; and 4) fear. This work was validated using an online study, which investigates how well the emotions expressed by Gertie are recognized by human audiences. In total 84 participants were shown one video for each of the four emotions and they were asked to choose from a list of seven emotions, which they thought was displayed by Gertie. While joy and sadness were recognized very reliably with 81% and 88% of all people giving the correct answer, fear and surprise were more commonly misinterpreted as surprise and disgust. However, all emotions were recognized above the chance level percentage of 14%.

Cognitive products integrate cognitive functionalities such as perceiving the environment, learning and reasoning from knowledge models that are created through the combination of a mechatronic system and advanced software algorithms. While the area of cognitive products is still in his infancy, we regard the safe and reliable performance fulfillment as one of the chal-lenging tasks for the research community to develop cognitive products that meet customer expectations. This paper presents how failure analysis can be integrated in the functional modeling process of cognitive products to in-crease their safety. For this purpose, we explores if current state failure anal-ysis methods are appropriate for analyzing cognitive products and where their weaknesses and strengths are. The cognitive coffee waiter is used as an illus-trative example to concretely show the limitations of these methods.

This paper presents a taxonomy of cognitive functions that supports formal functional modeling of cognitive technical systems (CTSs) and cognitive products. To date, there is little support for functional modeling of such systems and products even though their interdisciplinary complexity exceeds that of electro-mechanical products and makes modeling support in conceptual design even more important. The taxonomy of cognitive functions is based on literature research and consists of a set of cognitive capabilities on three hierarchical levels as well as a defined set of flows. Relationships among cognitive capabilities have been identified using WordNet, a lexical database of English. The application of the taxonomy is demonstrated through the example of a coffee robot waiter, which has been designed and prototyped in the research group of the authors. Through defining a common taxonomy of cognitive functions and flows, a common practice for functional modeling of cognitive products is defined thus supporting re-use of functional models. This creates the foundation for creating model-based design repositories for CTSs and cognitive products to support their future development.

This paper presents lessons learned from a project-based approach for teaching new cognitive product development to multi-disciplinary student teams. Within the class, interdisciplinary teams with students from mechanical engineering, electrical engineering and computer science are formed. To date, the class has been run five semesters using different themes and developing more than 15 new products. Cognitive products are tangible and durable things with cognitive capabilities that consist of a physical carrier system with embodied mechanics, electronics, microprocessors and software. The surplus value is created through cognitive capabilities, e.g. perceiving, learning and reasoning, that are enabled by flexible control loops and cognitive algorithms. Common product development processes are often not applicable to this high level of interdisciplinarity and require extension and refinement. Therefore, project-focused, technical lectures, balancing technology-push and market pull, are given and specifically tailored to cognitive product design. Additionally, software and hardware prototyping workshops and toolkits are provided in combination with the support of technical and process coaches to assist the teams. Attention is especially paid to giving students hands-on experience in developing their own cognitive product starting with the generation of ideas, turning the best idea into a product concept and finally building a functional and a design prototype. Lessons learned through the refinement of the product development process and methods used as well as from student questionnaires are presented and discussed

Cognitive products offer great potential to satisfy new user needs and desires through cognitive capabilities. This paper provides a review of related product types, e.g. mechatronic and "smart" products, and presents a definition of cognitive products derived from reviewing cognitive technical systems. Based on this definition, a framework for cognitive functions is developed and evaluated using a set of example cognitive products developed in student, project-based seminars. This framework enables the categorization of cognitive products through their functions and gives insight into future research directions including developing design catalogs for developing cognitive products.