Towards a Design Process for Computer-Aided Biomimetics.

biomimetics Communication Towards a Design Process for Computer-Aided Biomimetics Ruben Kruiper 1, * ID , Julian F. V. Vincent 2 , Eitan Abraham 2 , Rupert C. Soar 3 , Ioannis Konstas 1 , Jessica Chen-Burger 1 and Marc P. Y. Desmulliez 2 ID 1 2 3 * Deparment of Mathematical and Computer Sciences, Heriot-Watt University, Edinburgh Campus, Edinburgh EH14 4AS, UK; (I.K.); (J.C.-B.) School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh Campus, Edinburgh EH14 4AS, UK; (J.F.V.V.); (E.A.); (M.P.Y.D.) School of Architecture Design and the Built Environment, Nottingham Trent University, 50 Shakespeare St, Nottingham NG1 4FQ, UK; Correspondence: ; Tel.: +44-131-449-5111 Received: 6 April 2018; Accepted: 17 June 2018; Published: 21 June 2018



 Abstract: Computer-Aided Biomimetics (CAB) tools aim to support the integration of relevant biological knowledge into biomimetic problem-solving processes. Specific steps of biomimetic processes that require support include the identification, selection and abstraction of relevant biological analogies. Existing CAB tools usually aim to support these steps by describing biological systems in terms of functions, although engineering functions do not map naturally to biological functions. Consequentially, the resulting static, functional view provides an incomplete understanding of biological processes, which are dynamic, cyclic and self-organizing. This paper proposes an alternative approach that revolves around the concept of trade-offs. The aim is to include the biological context, such as environmental characteristics, that may provide information crucial to the transfer of biological information to an engineering application. The proposed design process is exemplified by an illustrative case study. Keywords: Computer-Aided Biomimetics (CAB); Biologically Inspired Design (BID); biomimicry; biomimetics; bionics; design theory; innovation; invention; problem-solving 1. Introduction This article focuses on finding solutions for technical problems that are inspired by nature. To this end, the definition for biomimetics by Fayemi et al. is followed, namely “the interdisciplinary creative process between biology and technology, aiming to solve technospheric problems through abstraction, transfer and application of knowledge from biological models” [1]. Unlike in engineering, functionalities encountered in nature are often hierarchical, dynamical and rely on information embedded at various hierarchical levels. As a result, biomimetics remains adventitious and is not used as widely and often as it potentially could be [2]. Problem-driven biomimetics processes are scarcely automated and usually take between 6 and 18 months to get from a specific problem to a functional prototype [3,4]. To enable a more systematic application of biomimetics, computational tools are required that integrate large amounts of biological knowledge in a given framework amenable to a methodology suitable for engineering. The initial aim of this article is to elucidate the requirements for such computational tools: • Avoid a bias towards engineering terminology and engineering functions, in order to preserve contextual information that is present in biological terminology; see Sections 3.1 and 3.2. Biomimetics 2018, 3, 14; doi:10.3390/biomimetics3030014 www.mdpi.com/journal/biomimetics Biomimetics 2018, 3, 14 • • • 2 of 21 Extract structured, within-domain information from biology research papers to enable reliable information retrieval; see Section 3.3. Avoid subsequent automated mapping between the biological and engineering domains, taking into consideration the semantic distance between these domains; see Sections 3.3 and 3.4. Support the direct and indirect uses of various theoretical models for biomimetics to represent the information found in biological texts, hence a model-agnostic tool; see Section 3.4. Section 2 addresses the support that Computer-Aided Biomimetics (CAB) tools should provide following a literature review [1,5–9]. The search for biological information is thought to be most comprehensive and effective if biological texts are used [10]. Support for the identification and filtering of relevant texts is often provided by linking biological systems to other domains of knowledge through the concept of function. The notion of a function bridge between the biological and engineering domains is sometimes invoked [11]. However, as described in Section 3, the role of function in biomimetics is not always justified or appropriate, especially during the automated identification of relevant biological texts [10,12–16]. A reader who is aware of the limitations of a functional approach in CAB may skip Sections 2 and 3. These sections are intended to provide a theoretical background for the high-level requirements listed above. Finding a suitable bridge to transfer knowledge between the domains of biology and engineering is challenging and remains a current topic of research. The assumption is that a direct transfer is often not feasible. Instead multiple biological systems may inspire a compound solution; see Section 4 on iterative design [17,18]. Section 5 provides a brief overview of existing CAB tools [19–21]. Shortcomings of these systems include their bias towards engineering terminology and the partial omission of systemic context implied in biological terms. The secondary aim of this article is to present progress towards an alternative approach to CAB that satisfies the requirements listed above. Section 6 introduces this approach in which the iterative search for relevant biological information is initially guided by a trade-off between two (or more) high-level features of a system. The purpose of a computational tool is to reduce the effort and time required to perform the proposed approach to biomimetics. In the proposed approach, this reduction is achieved by enabling a user to process more relevant raw biological information in a short amount of time. The task of a CAB tool that meets the requirements listed above is then to (1) present the important concepts and relations found verbatim in biological information sources and (2) improve retrieval by extracting trade-offs. 2. The Problem of Finding Relevant Biological Systems Figure 1 shows the difference between generic problem-solving as presented by Massey and Wallace, and taking inspiration from nature when solving a problem [22]. Both processes are depicted as a series of steps to get from a problem to a solution. Engineers who want to solve a biomimetic problem require support during the three steps that utilize information from the biological domain, as well as the two transfer steps between the biological and engineering domains. This is because most engineers know little biology or characteristics of animals and plants. A plethora of biomimetic design methods have therefore been proposed to provide this support [9]. Biomimetic (...truncated)