Article
Engineering, Mechanical
Qihao Zhu, Xinyu Zhang, Jianxi Luo
Summary: Biological systems have evolved for millions of years and their features can inspire and benefit technical problem-solving in modern industries. Bio-inspired design (BID) is a design method that utilizes these features, but the gap between biology and engineering hinders its effectiveness. This paper proposes a generative design approach based on a language model to retrieve and map biological analogies, generating BID concepts in natural language.
JOURNAL OF MECHANICAL DESIGN
(2023)
Article
Engineering, Industrial
Mingdong Li, Shanhe Lou, Yicong Gao, Hao Zheng, Bingtao Hu, Jianrong Tan
Summary: Conceptual design is a crucial stage in new product development, and the function-behaviour-structure framework is adopted to aid designers in searching and generating conceptual solutions. Computer-aided methods within this framework facilitate cognitive activities and propose a cerebellar operant conditioning-inspired approach to solve the mapping process from behaviours to structures. A modularised constraint satisfaction neural network is constructed, inspired by the cerebellar structure, to determine the satisfiability of design problems and generate conceptual solutions by clustering embedded nodes. This approach imitates design constraint-driven operant conditioning, reducing design iterations and avoiding combinatorial explosions in conceptual design.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Chemistry, Multidisciplinary
Ze Bian, Shijian Luo, Fei Zheng, Liuyu Wang, Ping Shan
Summary: This study focuses on assisting designers in bionic semantic reasoning in product biologically inspired design. Through experiments and the use of a pretraining model, semantic similarities between product description sentences and biological sentences were calculated to provide designers with efficient bionic reasoning choices.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Artificial Intelligence
Sweta Kumari, V. Srinivasa Chakravarthy
Summary: This study proposes an image classification model based on attentional search, which integrates information through a sequence of saccades and makes classification decisions. The model is trained using deep Q-learning and is evaluated on multiple datasets, demonstrating superior performance.
Article
Chemistry, Multidisciplinary
Junlei Zhang, Runhua Tan
Summary: This paper proposes an effective process of radical concept generation, which involves identifying radical technology opportunities, determining the search direction of cross-domain knowledge, selecting appropriate cross-domain knowledge, and forming radical concepts using the cross-domain knowledge as inspirations. This method can reduce risks and costs caused by uncertainty and has the potential to foster research on radical innovation and knowledge-based innovation.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Jun Li, Wu Zhao, Kai Zhang, Miao Yu, Xin Guo
Summary: Concept design is crucial in product design as it aims to create conceptual schemes that meet various constraints. One of the challenges is to avoid overlap and interference in the space of functions and structures while maximizing the use of workspace and achieving product functionality. This study proposes a space layout design model that combines function-based spatial planning and structure dynamic deployment to address this issue. The model was demonstrated through a case study of a machine for electrical insulator detection, showing its feasibility.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Computer Science, Interdisciplinary Applications
Qihao Zhu, Jianxi Luo
Summary: This study explores the application of natural language generation (NLG) technique in automated early stage design concept generation. The generative pretrained transformer (GPT) is used to leverage knowledge and reasoning from textual data and transform them into understandable new concepts. Experimental results show good performance in generating novel and useful concepts.
JOURNAL OF COMPUTING AND INFORMATION SCIENCE IN ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Guo-Niu Zhu, Jin Ma, Jie Hu
Summary: Biological inspiration evaluation is crucial in biologically inspired design, but the interdisciplinary nature of the assessment presents challenges in objectively evaluating under uncertain conditions. This study proposes a fuzzy rough number extended multi-criteria group decision-making approach to address the subjectivity and uncertainty in biological inspiration evaluation, which is validated to be superior through experimental results and comparative analysis.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2021)
Article
Chemistry, Multidisciplinary
Peng Zhang, Xindi Li, Zifeng Nie, Fei Yu, Wei Liu
Summary: This paper introduces a new design method that combines bio-inspired design and trimming method to enhance design innovation. The method includes trimming analysis, keyword search for biological prototypes, and fuzzy comprehensive evaluation. Finally, the biological solution is transformed into an engineering design scheme through a resource derivation process.
APPLIED SCIENCES-BASEL
(2021)
Article
Chemistry, Multidisciplinary
Ying Xu, Saeed Afshar, Runchun Wang, Gregory Cohen, Chetan Singh Thakur, Tara Julia Hamilton, Andre van Schaik
Summary: A biologically inspired sound localisation system using the CAR-FAC cochlear model was proposed for reverberant environments, which preprocesses binaural signals and analyses them using cross-correlation and a neural network for localization. The system's performance with the CAR-FAC nonlinearity was investigated and compared with other models to verify its effectiveness in sound localisation tasks.
APPLIED SCIENCES-BASEL
(2021)
Article
Mathematics, Interdisciplinary Applications
N. Biocca, P. J. Blanco, D. E. Caballero, J. M. Gimenez, G. E. Carr, S. A. Urquiza
Summary: This work proposes a technique for simultaneous mesh optimization and motion characterization inspired by soft tissues, with adaptive mechanisms like growth and remodeling to attain optimal configuration. Through a series of consecutive nonlinear optimization steps, the algorithm can achieve optimized configuration. The performance of the proposed method is demonstrated through 2D and 3D numerical experiments, showing satisfactory results.
COMPUTATIONAL MECHANICS
(2022)
Article
Acoustics
Tianlong Ma, Gang Qiao, Songzuo Liu, Suleman Mazhar, Naihua Zheng, Chenyu Pan
Summary: This article introduces a biologically inspired underwater acoustic communication method using discrete cosine transform (DCT) of cetacean sounds and watermark technology to achieve reliable underwater communication. The feasibility of the method is verified through simulations and experiments, and good communication rates and low bit error rates are achieved.
Review
Surgery
John C. Alverdy
Summary: As new stapling devices with robotic adaptations, tri-staple technology, preloaded reinforcement materials, etc. enter the market, there is a need for a fresh understanding of their mechanisms of action. While explaining the mechanical features of these devices has received much attention, little to no attention has been paid to understanding the biological response of intestinal anastomosis to the variations in their use and design. This perspective piece reviews the different aspects of gastrointestinal stapling in the context of emerging technology and highlights the gaps in knowledge regarding the effect of gastrointestinal stapling on healing biology.
AMERICAN JOURNAL OF SURGERY
(2023)
Article
Computer Science, Artificial Intelligence
Zhenshan Bing, Amir Ei Sewisy, Genghang Zhuang, Florian Walter, Fabrice O. Morin, Kai Huang, Alois Knoll
Summary: In this article, a computational HDC network consistent with neurophysiological findings concerning biological HDCs is proposed and implemented in robotic navigation tasks. The network represents the directional heading relying solely on angular velocity input and demonstrates excellent performance in accuracy and real-time capability through extensive simulations and real-world experiments.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Chemistry, Multidisciplinary
Michael Elwert, Manuel Ramsaier, Boris Eisenbart, Ralf Stetter, Markus Till, Stephan Rudolph
Summary: This paper presents a study on the integration of an integrated function modeling framework and a graph-based design language framework in engineering. The use of graph-based design languages allows for automated design and evaluation processes, as well as the generation of numerous viable design alternatives. The paper also expands the applicability of these frameworks into the domain of product functions and introduces a function analysis tool for assessing component reliabilities and the importance of processes in a technical system.
APPLIED SCIENCES-BASEL
(2022)