Fig 6 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Aquatic Creatures Generated by GPT-4 GPT-4 successfully generated variations of aquatic creatures automatically using primitives from the PyVista package in Python.
Source publication
The advancement of Large Language Models (LLMs), including GPT-4, provides exciting new opportunities for generative design. We investigate the application of this tool across the entire design and manufacturing workflow. Specifically, we scrutinize the utility of LLMs in tasks such as: converting a text-based prompt into a design specification, tr...
Contexts in source publication
Context 1
... to the library's documentation, GPT-4 is able to automatically assemble a functional python program using PyVista's primitive functions. We asked GPT-4 to use PyVista's primitives to model several variations of a fish, including specific bioinspirations such as goldfish, a manta ray, and a loach ( Figure 6). GPT-4 successfully selected and scaled an appropriate set of primitives for each example, and provided sound bio-inspired rationale for its decisions. ...
Context 2
... library. It chooses L-BFGS-B, which is a reasonable, standard, and easily accessible (though not state-of-the-art) solver for continuous valued problems. It does not, however, provide gradients that can expedite the computation unless prompted for them. We explicitly prompt GPT-4 to provide the gradients (Fig. 55) and visualize the results in Fig. 56. Generally speaking, the unoptimized approach on GPT-4's part is an issue w.r.t. performance, as not all users will be intimately familiar with all (or perhaps any) optimization libraries, and they may not realize that by providing additional information (e.g. gradients), the computation can be expedited. GPT-4 also does not elect to ...
Context 3
... the user verifies the accuracy of the functions, we have GPT-4 output a python script that can minimize the cabinet's material cost with respect to a given volume constraint. The resulting code is shown in Figure 60, with renders of an optimized cabinet in Figure 61. ...
Context 4
... it can apply knowledge of aspects of a problem to new problems in familiar domains. The conversation in Figure 65 presents a brief example of GPT-4 being queried about a novel invention: the Fworp 2 . The Fworp is a robot car with a body made of silicone rubber. ...
Context 5
... While silicone is durable, it can also be prone to punctures and cuts. Depending on the environment, the surface may need regular inspections and repairs. (4) Cost: Depending on the specifics, manufacturing a vehicle out of silicone could potentially be more expensive than using more traditional materials. (... content omitted by authors ...) Fig. 65. Testing GPT-4's Domain Transfer Ability for Inverse Design. We query GPT-4 about the Fworp, an imaginary invention for which it has no direct knowledge about. Despite this, GPT-4 is able to transfer its knowledge about related concepts to propose design parameters and comparisons. However, some of its proposed parameters, such as load ...
Context 6
... this subsection, we consolidate and elaborate on the comprehensive LLM-assisted development process for a wood cabinet with minimal human intervention. Throughout this process, GPT-4 played major roles in generating the 3D design, facilitating part sourcing based on the design specifications, generating machine-readable manufacturing instructions, providing human-readable assembly guidance, and conducting final performance evaluations ( Figure 66). This holistic approach highlights the extensive capabilities of GPT-4 in automating and optimizing various stages of this product development process. ...
Context 7
... including tasks such as adding a door and a handle (Section 4.2.1) or adjusting the number of shelves (Section 4.3). GPT-4 even integrated prefabricated elements like brackets into the cabinet design (Section 4.2.2). We ultimately decided to fabricate the version of the cabinet presented in Figure 7 with one shelf supported by four brackets. Fig. 66. The Cabinet Design Process Using GPT-4. We showcase the comprehensive development process for a wood cabinet, highlighting GPT-4's roles in 3D design generation, part sourcing, manufacturing instructions, assembly guidance, and performance evaluations. Now let's fabricate this cabinet using wood. I have a universal laser system with a ...
Context 8
... me modify the code accordingly: from McMaster. Each suggestion was accompanied by a concise description of the item (Figure 67 and Section 6.3). Furthermore, in order to realize the subtractive manufacturing approach, GPT-4 successfully generated precise 2D DXF files for laser cutting the wood pieces. ...
Context 9
... this size was too small to support the original cabinet design. As we presented this constraint to GPT-4, it successfully scaled down the entire design to ensure compatibility with the available wood pieces (Figure 68). Given that our laser cutter is capable of engraving patterns onto the cutting materials, we took advantage of this feature and requested GPT-4 to generate Python code for producing DXF files that could engrave words onto the side board of the cabinet. ...
Context 10
... suggested Quick-Set Epoxy, 3M DP100, 1.64 FL. oz. Cartridge as the optimal glue for our purposes (Figure 69). Additionally, GPT-4 provided detailed treatment guidance and step-by-step procedures for the assembly, ensuring a smooth and successful construction process (Figure 69). ...
Context 11
... as the optimal glue for our purposes (Figure 69). Additionally, GPT-4 provided detailed treatment guidance and step-by-step procedures for the assembly, ensuring a smooth and successful construction process (Figure 69). The final fabricated cabinet is showcased in Figure 70, offering three different view angles. ...
Context 12
... 76, we prompt GPT-4 to create a Component class whose instances store the geometry, mass, position, and orientation attributes of the corresponding OpenJSCAD primitives. Component instances also have distinct names to represent URDF links. We additionally prompt GPT-4 to generate helper functions for placing instances with different geometries in Fig. 76. This framework allows GPT-4 to generate a function that places components in terms of absolute coordinate positions and orientations and to replicate the Python equivalent of the OpenJSCAD, as shown in Fig. ...
Similar publications
Large language models (LLMs) have revolutionized natural language processing with their exceptional capabilities. However, deploying LLMs on resource-constrained edge devices presents significant challenges due to computational limitations, memory constraints, and edge hardware heterogeneity. This survey summarizes recent developments in edge LLMs...
Citations
... The advent of generative artificial intelligence (GenAI) has ushered a momentum of innovation in the field of additive manufacturing (AM), where the boundaries between the physical and digital worlds blur, giving rise to what has been termed Industry 5.0 [1][2][3]. At the forefront of this transformation are Large Language Models (LLMs), monumental advancements in natural language processing that have ignited a paradigm shift in how we approach design and manufacturing, particularly within the field of engineering, biomedicine, and biotechnological applications [4][5][6]. In recent years, the development and proliferation of LLMs, such as GPT-4, Gemini, Llama, and Microsoft Co-Pilot, have unleashed unprecedented capabilities for understanding and generating human-like text and image [7][8][9][10]. ...
... GPT-4 has demonstrated its ability to formulate design spaces, set objectives, and define constraints. It can also select suitable search algorithms for given problems, highlighting its utility as a foundational component in creating inverse design systems (Makatura et al., 2023). ...
In “AI-Driven Architecture: Pioneering the Digital Frontier”, explore how artificial intelligence is reshaping architecture and urban design, pushing the limits of what’s possible. From AI-supported design processes and evolving transparency in algorithms to policy impacts on smart transportation and new paradigms in architectural education, this book delves into the intersection of technology and creativity. Discover how AI is integrated into design processes, drives innovations in conceptual design, and even re-imagines the concept of infinity in architecture. Featuring case studies, thought-provoking insights, and practical examples, this book offers a comprehensive guide to the transformative power of AI in shaping the built environment.
... With the rise of groundbreaking tools such as ChatGPT, generative artificial intelligence (AI) has attracted significant attention, revolutionizing how AI is applied in various fields, including engineering design [1], [2]. This trend has seen the adoption of AI techniques in numerous design activities such as topology optimization, material design, design synthesis, and product design [3], [4], [5]. ...
... Current literature throws up ideas on utilizing LLMs, e.g. [31], or VFMs, e.g., [28][29][30]32], in the industrial domain; little is known about how to enable VFM to perform effectively in specific use cases. Besides having suitable datasets, training with the data demands specific strategies. ...
In recent years, the upstream of Large Language Models (LLM) has also encouraged the computer vision community to work on substantial multimodal datasets and train models on a scale in a self-/semi-supervised manner, resulting in Vision Foundation Models (VFM), as, e.g., Contrastive Language-Image Pre-training (CLIP). The models generalize well and perform outstandingly on everyday objects or scenes, even on downstream tasks, tasks the model has not been trained on, while the application in specialized domains, as in an industrial context, is still an open research question. Here, fine-tuning the models or transfer learning on domain-specific data is unavoidable when objecting to adequate performance. In this work, we, on the one hand, introduce a pipeline to generate the Industrial Language-Image Dataset (ILID) based on web-crawled data; on the other hand, we demonstrate effective self-supervised transfer learning and discussing downstream tasks after training on the cheaply acquired ILID, which does not necessitate human labeling or intervention. With the proposed approach, we contribute by transferring approaches from state-of-the-art research around foundation models, transfer learning strategies, and applications to the industrial domain.
... With the development of generative AI, in particular large language models (LLMs), design has taken a new shift in many ways. Broadly situated, the LLMs potential in design lies in enabling five key tasks: converting text prompts to design specifications, converting designs to manufacturing instructions, creating design spaces and variations, calculating performance, and exploring performance-based design solutions [14]. Additionally, in the early design phases, a student designer in need of better representations and appreciations of an imagined product can choose to visualise concepts in high fidelity without the need for prototyping or expertise in visualisation techniques, such as rendering, and to vary the aesthetics of those concept designs through chatbot-type interfaces. ...
Conceptual design is an essential stage in the design process, and its ultimate success largely depends on designers’ creativity. Both physical and digital prototypes are commonly adopted by designers to support ideation and creativity, providing intuitive perception and rapid iteration, respectively. In recent advancements, large-scale generation models are able to offer data-enabled creativity support by generating high-quality solutions comparable to human designers. This opens up an imaginary space for designers and brings new possibilities for design tools. In this study, we proposed a hybrid prototype method that synergistically combines physical models and generative artificial intelligence (AI) in the conceptual design stage. Correspondingly, we developed a hybrid prototype system to implement the proposed method. We conducted a comparative user study with 45 designers who completed a design task using the physical prototype method, standalone generative AI, and the hybrid prototype method, respectively. Our results verified the effectiveness of the hybrid prototype method and investigated its mechanism in supporting creativity. Finally, we discussed the application value and optimisation space of the hybrid prototype method.