arXiv subjects

Hyunmin Cheong

Publications and source records attributed to Hyunmin Cheong.

14 recordsLinked to original sources

arXiv2025

Transformer-Based Interfaces for Mechanical Assembly Design: A Gear Train Case Study

Generative artificial intelligence (AI), particularly transformer-based models, presents new opportunities for automating and augmenting engineering design workflows. However, effectively integrating these models into interactive tools requires careful interface design that leverages their unique capabilities. This paper introduces a transformer model tailored for gear train assembly design, paired with two novel interaction modes: Explore and Copilot. Explore Mode uses probabilistic sampling to generate and evaluate diverse design alternatives, while Copilot Mode utilizes autoregressive prediction to support iterative, context-aware refinement. These modes emphasize key transformer properties (sequence-based generation and probabilistic exploration) to facilitate intuitive and efficient human-AI collaboration. Through a case study, we demonstrate how well-designed interfaces can enhance engineers' ability to balance automation with domain expertise. A user study shows that Explore Mode supports rapid exploration and problem redefinition, while Copilot Mode provides greater control and fosters deeper engagement. Our results suggest that hybrid workflows combining both modes can effectively support complex, creative engineering design processes.

cs.HC↗

arXiv2025

Physics-based simulation ontology: an ontology to support modelling and reuse of data for physics-based simulation

The current work presents an ontology developed for physics-based simulation in engineering design, called Physics-based Simulation Ontology (PSO). The purpose of the ontology is to assist in modelling the physical phenomenon of interest in a veridical manner, while capturing the necessary and reusable information for physics-based simulation solvers. The development involved extending an existing upper ontology, Basic Formal Ontology (BFO), to define lower-level terms of PSO. PSO has two parts: PSO-Physics, which consists of terms and relations used to model physical phenomena based on the perspective of classical mechanics involving partial differential equations, and PSO-Sim, which consists of terms used to represent the information artefacts that are about the physical phenomena modelled with PSO-Physics. The former terms are used to model the physical phenomenon of interest independent of solver-specific interpretations, which can be reused across different solvers, while the latter terms are used to instantiate solver-specific input data. A case study involving two simulation solvers was conducted to demonstrate this capability of PSO. Discussion around the benefits and limitations of using BFO for the current work is also provided, which should be valuable for any future work that extends an existing upper ontology to develop ontologies for engineering applications.

cs.AI↗

arXiv2025

Configuration Design of Mechanical Assemblies using an Estimation of Distribution Algorithm and Constraint Programming

A configuration design problem in mechanical engineering involves finding an optimal assembly of components and joints that realizes some desired performance criteria. Such a problem is a discrete, constrained, and black-box optimization problem. A novel method is developed to solve the problem by applying Bivariate Marginal Distribution Algorithm (BMDA) and constraint programming (CP). BMDA is a type of Estimation of Distribution Algorithm (EDA) that exploits the dependency knowledge learned between design variables without requiring too many fitness evaluations, which tend to be expensive for the current application. BMDA is extended with adaptive chi-square testing to identify dependencies and Gibbs sampling to generate new solutions. Also, repair operations based on CP are used to deal with infeasible solutions found during search. The method is applied to a vehicle suspension design problem and is found to be more effective in converging to good solutions than a genetic algorithm and other EDAs. These contributions are significant steps towards solving the difficult problem of configuration design in mechanical engineering with evolutionary computation.

cs.NE↗

arXiv2025

Optimal Design of Continuum Robots with Reachability Constraints

While multi-joint continuum robots are highly dexterous and flexible, designing an optimal robot can be challenging due to its kinematics involving curvatures. Hence, the current work presents a computational method developed to find optimal designs of continuum robots given reachability constraints. First, we leverage both forward and inverse kinematic computations to perform reachability analysis in an efficient yet accurate manner. While implementing inverse kinematics, we also integrate torque minimization at joints such that robot configurations with the minimum actuator torque required to reach a given workspace could be found. Lastly, we apply an estimation of distribution algorithm (EDA) to find optimal robot dimensions while considering reachability, where the objective function could be the total length of the robot or the actuator torque required to operate the robot. Through three application problems, we show that the EDA is superior to a genetic algorithm (GA) in finding better solutions within a given number of iterations, as the objective values of the best solutions found by the EDA are 4-15\% lower than those found by the GA.

cs.RO↗

arXiv2025

mPOLICE: Provable Enforcement of Multi-Region Affine Constraints in Deep Neural Networks

Deep neural networks are increasingly used in safety-critical domains such as robotics and scientific modeling, where strict adherence to output constraints is essential. Methods like POLICE, which are tailored for single convex regions, face challenges when extended to multiple disjoint regions, often leading to constraint violations or unwanted affine behavior across regions. This paper proposes mPOLICE, a new approach that generalizes POLICE to provably enforce affine constraints over multiple disjoint convex regions. At its core, mPOLICE assigns distinct neuron activation patterns to each constrained region, enabling localized affine behavior and avoiding unintended generalization. This is implemented through a layer-wise optimization of the network parameters. Additionally, we introduce a training algorithm that incorporates mPOLICE into conventional deep learning pipelines, balancing task-specific performance with constraint enforcement using periodic sign pattern enforcement. We validate the flexibility and effectiveness of mPOLICE through experiments across various applications, including safety-critical reinforcement learning, implicit 3D shape representation with geometric constraints, and fluid dynamics simulations with boundary condition enforcement. Importantly, mPOLICE incurs no runtime overhead during inference, making it a practical and reliable solution for constraint handling in deep neural networks.

cs.LG↗

arXiv2025

e-SimFT: Alignment of Generative Models with Simulation Feedback for Pareto-Front Design Exploration

Deep generative models have recently shown success in solving complex engineering design problems where models predict solutions that address the design requirements specified as input. However, there remains a challenge in aligning such models for effective design exploration. For many design problems, finding a solution that meets all the requirements is infeasible. In such a case, engineers prefer to obtain a set of Pareto optimal solutions with respect to those requirements, but uniform sampling of generative models may not yield a useful Pareto front. To address this gap, we introduce a new framework for Pareto-front design exploration with simulation fine-tuned generative models. First, the framework adopts preference alignment methods developed for Large Language Models (LLMs) and showcases the first application in fine-tuning a generative model for engineering design. The important distinction here is that we use a simulator instead of humans to provide accurate and scalable feedback. Next, we propose epsilon-sampling, inspired by the epsilon-constraint method used for Pareto-front generation with classical optimization algorithms, to construct a high-quality Pareto front with the fine-tuned models. Our framework, named e-SimFT, is shown to produce better-quality Pareto fronts than existing multi-objective alignment methods.

cs.LG↗

arXiv2024

Optimal design of frame structures with mixed categorical and continuous design variables using the Gumbel-Softmax method

In optimizing real-world structures, due to fabrication or budgetary restraints, the design variables may be restricted to a set of standard engineering choices. Such variables, commonly called categorical variables, are discrete and unordered in essence, precluding the utilization of gradient-based optimizers for the problems containing them. In this paper, incorporating the Gumbel-Softmax (GSM) method, we propose a new gradient-based optimizer for handling such variables in the optimal design of large-scale frame structures. The GSM method provides a means to draw differentiable samples from categorical distributions, thereby enabling sensitivity analysis for the variables generated from such distributions. The sensitivity information can greatly reduce the computational cost of traversing high-dimensional and discrete design spaces in comparison to employing gradient-free optimization methods. In addition, since the developed optimizer is gradient-based, it can naturally handle the simultaneous optimization of categorical and continuous design variables. Through three numerical case studies, different aspects of the proposed optimizer are studied and its advantages over population-based optimizers, specifically a genetic algorithm, are demonstrated.

cs.CE↗

arXiv2024

A low order, torsion deformable spatial beam element based on the absolute nodal coordinate formulation and Bishop frame

Heretofore, the Serret-Frenet frame has been the ubiquitous choice for analyzing the elastic deformations of beam elements. It is well known that this frame is undefined at the inflection points and straight segments of the beam where its curvature is zero, leading to singularities and errors in their numerical analysis. On the other hand, there exists a lesser-known frame called Bishop which does not have the caveats of the Serret-Frenet frame and is well-defined everywhere along the beam centerline. Leveraging the Bishop frame, in this paper, we propose a new spatial, singularity-free low-order beam element based on the absolute nodal coordinate formulation for both small and large deformation applications. This element, named ANCF14, has a constant mass matrix and can capture longitudinal, transverse (bending) and torsional deformations. It is a two-noded element with 7 degrees of freedom per node, which are global nodal coordinates, nodal slopes and their cross-sectional rotation about the centerline. The newly developed element is tested through four complex benchmarks. Comparing the ANCF14 results with theoretical and numerical results provided in other studies confirms the efficiency and accuracy of the proposed element.

cs.CE↗

arXiv2024

Design optimization of dynamic flexible multibody systems using the discrete adjoint variable method

The design space of dynamic multibody systems (MBSs), particularly those with flexible components, is considerably large. Consequently, having a means to efficiently explore this space and find the optimum solution within a feasible timeframe is crucial. It is well known that for problems with several design variables, sensitivity analysis using the adjoint variable method extensively reduces the computational costs. This paper presents the novel extension of the discrete adjoint variable method to the design optimization of dynamic flexible MBSs. The extension involves deriving the adjoint equations directly from the discrete, rather than the continuous, equations of motion. This results in a system of algebraic equations that is computationally less demanding to solve compared to the system of differential algebraic equations produced by the continuous adjoint variable method. To describe the proposed method, it is integrated with a numerical time-stepping algorithm based on geometric variational integrators. The developed technique is then applied to the optimization of MBSs composed of springs, dampers, beams and rigid bodies, considering both geometrical (e.g., positions of joints) and non-geometrical (e.g., mechanical properties of components) design variables. To validate the developed methods and show their applicability, three numerical examples are provided.

math.OC↗

arXiv2024

Physics Context Builders: A Modular Framework for Physical Reasoning in Vision-Language Models

Physical reasoning remains a significant challenge for Vision-Language Models (VLMs). This limitation arises from an inability to translate learned knowledge into predictions about physical behavior. Although continual fine-tuning can mitigate this issue, it is expensive for large models and impractical to perform repeatedly for every task. This necessitates the creation of modular and scalable ways to teach VLMs about physical reasoning. To that end, we introduce Physics Context Builders (PCBs), a modular framework where specialized smaller VLMs are fine-tuned to generate detailed physical scene descriptions. These can be used as physical contexts to enhance the reasoning capabilities of larger VLMs. PCBs enable the separation of visual perception from reasoning, allowing us to analyze their relative contributions to physical understanding. We perform experiments on CLEVRER and on Falling Tower, a stability detection dataset with both simulated and real-world scenes, to demonstrate that PCBs provide substantial performance improvements, increasing average accuracy by up to 13.8% on complex physical reasoning tasks. Notably, PCBs also show strong Sim2Real transfer, successfully generalizing from simulated training data to real-world scenes.

cs.CV↗

arXiv2024

Optimal Design of Vehicle Dynamics Using Gradient-Based, Mixed-Fidelity Multidisciplinary Optimization

In automotive engineering, designing for optimal vehicle dynamics is challenging due to the complexities involved in analysing the behaviour of a multibody system. Typically, a simplified set of dynamics equations for only the key bodies of the vehicle such as the chassis and wheels are formulated while reducing their degrees of freedom. In contrast, one could employ high-fidelity multibody dynamics simulation and include more intricate details such as the individual suspension components while considering full degrees of freedom for all bodies; however, this is more computationally demanding. Also, for gradient-based design optimization, computing adjoints for different objective functions can be more challenging for the latter approach, and often not feasible if an existing multibody dynamics solver is used. We propose a mixed-fidelity multidisciplinary approach, in which a simplified set of dynamics equations are used to model the whole vehicle while incorporating a high-fidelity multibody suspension module as an additional coupled discipline. We then employ MAUD (modular analysis and unified derivatives) to combine analytical derivatives based on the dynamics equations and finite differences obtained using an existing multibody solver. Also, we use a collocation method for time integration, which solves for both the system trajectory and optimal design variables simultaneously. The benefits of our approach are shown in an experiment conducted to find optimal vehicle parameters that optimize ride comfort and driving performance considering vertical vehicle dynamics.

eess.SY↗

arXiv2024

Deep Generative Model for Mechanical System Configuration Design

Generative AI has made remarkable progress in addressing various design challenges. One prominent area where generative AI could bring significant value is in engineering design. In particular, selecting an optimal set of components and their interfaces to create a mechanical system that meets design requirements is one of the most challenging and time-consuming tasks for engineers. This configuration design task is inherently challenging due to its categorical nature, multiple design requirements a solution must satisfy, and the reliance on physics simulations for evaluating potential solutions. These characteristics entail solving a combinatorial optimization problem with multiple constraints involving black-box functions. To address this challenge, we propose a deep generative model to predict the optimal combination of components and interfaces for a given design problem. To demonstrate our approach, we solve a gear train synthesis problem by first creating a synthetic dataset using a grammar, a parts catalogue, and a physics simulator. We then train a Transformer using this dataset, named GearFormer, which can not only generate quality solutions on its own, but also augment search methods such as an evolutionary algorithm and Monte Carlo tree search. We show that GearFormer outperforms such search methods on their own in terms of satisfying the specified design requirements with orders of magnitude faster generation time. Additionally, we showcase the benefit of hybrid methods that leverage both GearFormer and search methods, which further improve the quality of the solutions.

cs.AI↗

arXiv2024

DesignQA: A Multimodal Benchmark for Evaluating Large Language Models' Understanding of Engineering Documentation

This research introduces DesignQA, a novel benchmark aimed at evaluating the proficiency of multimodal large language models (MLLMs) in comprehending and applying engineering requirements in technical documentation. Developed with a focus on real-world engineering challenges, DesignQA uniquely combines multimodal data-including textual design requirements, CAD images, and engineering drawings-derived from the Formula SAE student competition. Different from many existing MLLM benchmarks, DesignQA contains document-grounded visual questions where the input image and input document come from different sources. The benchmark features automatic evaluation metrics and is divided into segments-Rule Comprehension, Rule Compliance, and Rule Extraction-based on tasks that engineers perform when designing according to requirements. We evaluate state-of-the-art models (at the time of writing) like GPT-4o, GPT-4, Claude-Opus, Gemini-1.0, and LLaVA-1.5 against the benchmark, and our study uncovers the existing gaps in MLLMs' abilities to interpret complex engineering documentation. The MLLMs tested, while promising, struggle to reliably retrieve relevant rules from the Formula SAE documentation, face challenges in recognizing technical components in CAD images, and encounter difficulty in analyzing engineering drawings. These findings underscore the need for multimodal models that can better handle the multifaceted questions characteristic of design according to technical documentation. This benchmark sets a foundation for future advancements in AI-supported engineering design processes. DesignQA is publicly available at: https://github.com/anniedoris/design_qa/.

cs.AI↗

arXiv2024

Elicitron: An LLM Agent-Based Simulation Framework for Design Requirements Elicitation

Requirements elicitation, a critical, yet time-consuming and challenging step in product development, often fails to capture the full spectrum of user needs. This may lead to products that fall short of expectations. This paper introduces a novel framework that leverages Large Language Models (LLMs) to automate and enhance the requirements elicitation process. LLMs are used to generate a vast array of simulated users (LLM agents), enabling the exploration of a much broader range of user needs and unforeseen use cases. These agents engage in product experience scenarios, through explaining their actions, observations, and challenges. Subsequent agent interviews and analysis uncover valuable user needs, including latent ones. We validate our framework with three experiments. First, we explore different methodologies for diverse agent generation, discussing their advantages and shortcomings. We measure the diversity of identified user needs and demonstrate that context-aware agent generation leads to greater diversity. Second, we show how our framework effectively mimics empathic lead user interviews, identifying a greater number of latent needs than conventional human interviews. Third, we showcase that LLMs can be used to analyze interviews, capture needs, and classify them as latent or not. Our work highlights the potential of using LLM agents to accelerate early-stage product development, reduce costs, and increase innovation.

cs.HC↗