Hasil untuk "Machine design and drawing"

Wang Jiaqi, Lan Yi, Chen Xiang

This study investigates AI integration in architectural education through a teaching experiment in Zhejiang University's 2024-25 grade three undergraduate design studio. Adopting a dual-module framework (20-hour AI training + embedded ethics discussions), the course introduced deep learning models, LLMs, AIGC, LoRA, and ComfyUI while maintaining the original curriculum structure, supported by dedicated technical instructors. Findings demonstrate the effectiveness of phased guidance, balanced technical-ethical approaches, and institutional support. The model improved students' digital skills and strategic cognition while addressing AI ethics, providing a replicable approach combining technical and critical learning in design education.

Zhen Zhang, Jun Hui Qiu, Jun Wei Zhang et al.

Reconfigurable intelligent surface (RIS) is a two-dimensional periodic structure integrated with a large number of reflective elements, which can manipulate electromagnetic waves in a digital way, offering great potentials for wireless communication and radar detection applications. However, conventional RIS designs highly rely on extensive full-wave EM simulations that are extremely time-consuming. To address this challenge, we propose a machine-learning-assisted approach for efficient RIS design. An accurate and fast model to predict the reflection coefficient of RIS element is developed by combining a multi-layer perceptron neural network (MLP) and a dual-port network, which can significantly reduce tedious EM simulations in the network training. A RIS has been practically designed based on the proposed method. To verify the proposed method, the RIS has also been fabricated and measured. The experimental results are in good agreement with the simulation results, which validates the efficacy of the proposed method in RIS design.

Souradip Poddar, Chia-Tung Ho, Ziming Wei et al.

Conventional AI-driven AMS design automation algorithms remain constrained by their reliance on high-quality datasets to capture underlying circuit behavior, coupled with poor transferability across architectures, and a lack of adaptive mechanisms. This work proposes HeaRT, a hierarchical circuit reasoning-based agentic framework for automation loops and a step toward adaptive, human-style design optimization. HeaRT consistently improves F1(subcircuits) by >= 13.5% and F1(loops) by >= 37.8% over few-shot prompting baselines across multiple LLM backbones on our 40-circuit AMS benchmark of flattened SPICE netlists, even as circuit complexity increases. Our experiments further show that HeaRT achieves >= 3x faster convergence in incremental design adaptation tasks under specification shifts across diverse optimization approaches, supporting both topology reconfiguration and sizing.

Soichiro Ide, Ayato Kanada, Yasutaka Nakashima et al.

Abstract Slewing cranes are useful for the handling work of heavy objects, such as in construction works. However, the load sway of the crane is a major problem, because it reduces safety and workability. Especially in large slewing cranes, the deflection of the boom causes vibrations in both the load and the boom, making maneuvering and control difficult. Therefore, a vibration suppression method considering boom deflection and sway of suspended load is important. This study proposes a feedback control method based on an inverse dynamics calculation, which aims to suppress suspended load vibration by allowing the suspended load to follow the target trajectory. In this paper, we describe an inverse dynamics calculation method that considers the boom as a rigid body. This calculation is used as a compensation factor for the suspended load position, and a feedback control method is proposed to track the suspended load position to the target trajectory. Some simulation of the operation using the proposed method is performed to verify the vibration suppression.

Peng Liu, Weiwei Zhang, Xuncheng Wu et al.

Vehicle accidents, particularly PV-PV collisions, result in significant property damage and driver injuries, causing substantial economic losses and health risks. Most existing studies focus on macro-level predictions, such as accident frequency, but lack detailed collision-level analysis, which limits the precision of severity prediction. This study investigates various accident-related factors, including environmental conditions, vehicle attributes, driver characteristics, pre-crash scenarios, and collision dynamics. Data from NHTSA’s CRSS and FARS datasets were integrated and balanced using random over-sampling and under-sampling techniques to address severity-level data imbalances. The mRMR algorithm was employed for feature selection to minimize redundancy and identify key features. Five advanced machine learning models were evaluated for severity prediction, with XGBoost achieving the best performance: 84.9% accuracy, 84.85% precision, 84.90% recall, and an F1-score of 84.87%. SHAP analysis was utilized to interpret the model and conduct a comprehensive analysis of accident features, including their importance, dependencies, and combined effects on severity prediction. This study achieved high accuracy in predicting accident severity across all levels in PV-PV collisions. Moreover, by integrating the SHAP model interpretation method, we conducted detailed feature analysis at global, local, and individual case levels, thereby filling the gap in PV-PV accident severity prediction and feature analysis.

Chew Kuew Wai, Taha Sadeq, Lee Cheun Hau

The energy storage system (ESS) plays a crucial role in electric vehicles (EVs), impacting their performance and efficiency. While batteries are the standard choice for energy storage, they come with drawbacks like low power density and limited life cycles, which can hinder pure battery electric vehicles (PBEVs). To address these issues, a hybrid energy storage system (HESS) that combines a battery with a supercapacitor provides a more effective solution. The battery delivers consistent power, while the supercapacitor manages peak power demands and regenerative braking energy. This study proposes a new energy management strategy for the HESS, an advanced adaptive rule-based algorithm. The results of the standard rule-based and adaptive rule-based algorithms are used to verify the proposed control algorithm. The system was modeled in MATLAB/Simulink and evaluated across three driving cycles—UDDS, NYCC, and Japan1015—while varying states of charge for the supercapacitors. The findings indicate that the HESS significantly alleviates battery stress compared to a pure battery system, enhancing both efficiency and lifespan. Among the algorithms tested, the advanced adaptive rule-based algorithm yielded the best results, increasing the number of viable drive cycles.

Zainab Saadoon Naser, Hend Marouane, Ahmed Fakhfakh

Traffic congestion has increased significantly in today’s rapidly urbanizing world, influencing people’s daily lives. Traffic signal control systems (TSCSs) play an important role in alleviating congestion by optimizing traffic light timings and improving road efficiency. Yet traditional TSCSs neglected pedestrians, cyclists, and other non-monitored road users, degrading traffic signal optimization (TSO). Therefore, this framework proposes a multi-object-based traffic flow analysis and intensity estimation model for efficient TSO using Upper Confidence Bound Multi-agent Reinforcement Learning Cubic Spline Fuzzy Logic (UCB-MRL-CSFL). Initially, the real-time traffic videos undergo frame conversion and redundant frame removal, followed by preprocessing. Then, the lanes are detected; further, the objects are detected using Temporal Context You Only Look Once (TC-YOLO). Now, the object counting in each lane is carried out using the Cumulative Vehicle Motion Kalman Filter (CVMKF), followed by queue detection using Vehicle Density Mapping (VDM). Next, the traffic flow is analyzed by Feature Variant Optical Flow (FVOF), followed by traffic intensity estimation. Now, based on the siren flashlight colors, emergency vehicles are separated. Lastly, UCB-MRL-CSFL optimizes the Traffic Signals (TSs) based on the separated emergency vehicle, pedestrian information, and traffic intensity. Therefore, the proposed framework outperforms the other conventional methodologies for TSO by considering pedestrians, cyclists, and so on, with higher computational efficiency (94.45%).

Daisuke Shionoiri, Yukiko Osawa

Abstract Robot skin is essential for achieving physical human-robot interaction. In particular, a warm feeling with a soft tactile impression is necessary for a human-friendly robot design. Contact sensing for safe interaction is also important; however, adding sensors to the skin surface can compromise its soft, smooth feel and create uncomfortable tactile sensations for humans. The paper proposes the design of a fluid-based soft robot skin that can simultaneously give thermal stimulation and detect human contact. We tested a prototype of the robotic skin, confirming that the fluid-based control successfully enabled both thermal display and contact detection capabilities.

Daqian Shao, Ashkan Soleymani, Francesco Quinzan et al.

A common issue in learning decision-making policies in data-rich settings is spurious correlations in the offline dataset, which can be caused by hidden confounders. Instrumental variable (IV) regression, which utilises a key unconfounded variable known as the instrument, is a standard technique for learning causal relationships between confounded action, outcome, and context variables. Most recent IV regression algorithms use a two-stage approach, where a deep neural network (DNN) estimator learnt in the first stage is directly plugged into the second stage, in which another DNN is used to estimate the causal effect. Naively plugging the estimator can cause heavy bias in the second stage, especially when regularisation bias is present in the first stage estimator. We propose DML-IV, a non-linear IV regression method that reduces the bias in two-stage IV regressions and effectively learns high-performing policies. We derive a novel learning objective to reduce bias and design the DML-IV algorithm following the double/debiased machine learning (DML) framework. The learnt DML-IV estimator has strong convergence rate and $O(N^{-1/2})$ suboptimality guarantees that match those when the dataset is unconfounded. DML-IV outperforms state-of-the-art IV regression methods on IV regression benchmarks and learns high-performing policies in the presence of instruments.

Seetaram Maurya, Nishchal K. Verma

Various methods for designing input features have been proposed for fault recognition in rotating machines using one-dimensional raw sensor data. The available methods are complex, rely on empirical approaches, and may differ depending on the condition monitoring data used. Therefore, this article proposes a novel algorithm to design input features that unifies the feature extraction process for different time-series sensor data. This new insight for designing/extracting input features is obtained through the lens of histogram theory. The proposed algorithm extracts discriminative input features, which are suitable for a simple classifier to deep neural network-based classifiers. The designed input features are given as input to the classifier with end-to-end training in a single framework for machine conditions recognition. The proposed scheme has been validated through three real-time datasets: a) acoustic dataset, b) CWRU vibration dataset, and c) IMS vibration dataset. The real-time results and comparative study show the effectiveness of the proposed scheme for the prediction of the machine's health states.

Yuxuan Yin, Yu Wang, Boxun Xu et al.

Analog circuit design requires substantial human expertise and involvement, which is a significant roadblock to design productivity. Bayesian Optimization (BO), a popular machine learning based optimization strategy, has been leveraged to automate analog design given its applicability across various circuit topologies and technologies. Traditional BO methods employ black box Gaussian Process surrogate models and optimized labeled data queries to find optimization solutions by trading off between exploration and exploitation. However, the search for the optimal design solution in BO can be expensive from both a computational and data usage point of view, particularly for high dimensional optimization problems. This paper presents ADO-LLM, the first work integrating large language models (LLMs) with Bayesian Optimization for analog design optimization. ADO-LLM leverages the LLM's ability to infuse domain knowledge to rapidly generate viable design points to remedy BO's inefficiency in finding high value design areas specifically under the limited design space coverage of the BO's probabilistic surrogate model. In the meantime, sampling of design points evaluated in the iterative BO process provides quality demonstrations for the LLM to generate high quality design points while leveraging infused broad design knowledge. Furthermore, the diversity brought by BO's exploration enriches the contextual understanding of the LLM and allows it to more broadly search in the design space and prevent repetitive and redundant suggestions. We evaluate the proposed framework on two different types of analog circuits and demonstrate notable improvements in design efficiency and effectiveness.

McPhillips Marita, Tegtmeier Silke, Nikitina Tatjana

Using different interaction channels within open innovation partnerships holds the potential to enhance the chance of success in production sectors. However, our comprehension of how open innovation partnerships are affected by varying combinations of interaction channels, and how this corelates with their level of open innovation output, remains limited. There are discrepancies in the current literature regarding the individual and combined effects of open innovation interaction channels. Our study aims to resolve these inconsistencies by using a configurational perspective, which allows for the identification of multiple successful pathways. Employing fuzzy-set Qualitative Comparative Analysis (fsQCA) to a dataset of European open innovation partnerships in production sectors, we uncover specific combinations of interaction channels that explain high levels of innovation outcomes. Subsequently, we distinguish between two successful pathways. Notably, we observe that the relationship between interaction channels is causally complex, high engagement in open innovation may not guarantee favorable innovation outcomes. This finding highlights the intricate causal dynamics at play. Thus, our study is a significant step toward reconciling the disparate perspectives in the literature concerning the impact of interaction channels on open innovation output.

Soobin Jeon, Junehong Park, Dongmahn Seo

Autonomous driving technology requires rigorous testing and validation of perception, decision-making, and control algorithms to ensure safety and reliability. Although existing simulators and testing tools play critical roles in algorithm evaluation, they struggle to satisfy the demands of complex, real-time systems. This study proposes a re-simulation-based integrated analysis system designed to overcome these challenges by providing advanced visualization, algorithm-testing, re-simulation, and data-handling capabilities. The proposed system features a comprehensive visualization module for real-time analysis of diverse sensor data and ego vehicle information, offering intuitive insights to researchers. Additionally, it includes a flexible algorithm-testing framework that abstracts simulator-specific dependencies, enabling seamless integration and evaluation of algorithms in various scenarios. The system also introduces robust re-simulation capabilities, enhancing algorithm validation using iterative testing based on real-world or simulated sensor data. To address the computational demands of high-frequency sensor data, the system employs optimized data-handling mechanisms based on shared memory, thereby significantly reducing latency and improving scalability. The proposed system overcomes critical challenges faced by existing alternatives by providing a robust, efficient, and scalable solution for testing and validating autonomous-driving algorithms, ultimately accelerating the development of safe and reliable autonomous vehicles.

Conor Walsh, Alok Joshi

Sports betting's recent federal legalisation in the USA coincides with the golden age of machine learning. If bettors can leverage data to reliably predict the probability of an outcome, they can recognise when the bookmaker's odds are in their favour. As sports betting is a multi-billion dollar industry in the USA alone, identifying such opportunities could be extremely lucrative. Many researchers have applied machine learning to the sports outcome prediction problem, generally using accuracy to evaluate the performance of predictive models. We hypothesise that for the sports betting problem, model calibration is more important than accuracy. To test this hypothesis, we train models on NBA data over several seasons and run betting experiments on a single season, using published odds. We show that using calibration, rather than accuracy, as the basis for model selection leads to greater returns, on average (return on investment of $+34.69\%$ versus $-35.17\%$) and in the best case ($+36.93\%$ versus $+5.56\%$). These findings suggest that for sports betting (or any probabilistic decision-making problem), calibration is a more important metric than accuracy. Sports bettors who wish to increase profits should therefore select their predictive model based on calibration, rather than accuracy.

Jiaao Chen, Aston Zhang, Xingjian Shi et al.

Parameter-efficient fine-tuning aims to achieve performance comparable to fine-tuning, using fewer trainable parameters. Several strategies (e.g., Adapters, prefix tuning, BitFit, and LoRA) have been proposed. However, their designs are hand-crafted separately, and it remains unclear whether certain design patterns exist for parameter-efficient fine-tuning. Thus, we present a parameter-efficient fine-tuning design paradigm and discover design patterns that are applicable to different experimental settings. Instead of focusing on designing another individual tuning strategy, we introduce parameter-efficient fine-tuning design spaces that parameterize tuning structures and tuning strategies. Specifically, any design space is characterized by four components: layer grouping, trainable parameter allocation, tunable groups, and strategy assignment. Starting from an initial design space, we progressively refine the space based on the model quality of each design choice and make greedy selection at each stage over these four components. We discover the following design patterns: (i) group layers in a spindle pattern; (ii) allocate the number of trainable parameters to layers uniformly; (iii) tune all the groups; (iv) assign proper tuning strategies to different groups. These design patterns result in new parameter-efficient fine-tuning methods. We show experimentally that these methods consistently and significantly outperform investigated parameter-efficient fine-tuning strategies across different backbone models and different tasks in natural language processing.

Minghan Chu, Weicheng Qian

Turbulent flows are of central importance across applications in science and engineering problems. For design and analysis, scientists and engineers use Computational Fluid Dynamics (CFD) simulations using turbulence models. Turbulent models are limited approximations, introducing epistemic uncertainty in CFD results. For reliable design and analysis, we require quantification of these uncertainties. The Eigenspace Perturbation Method (EPM) is the preeminent physics based approach for turbulence model UQ, but often leads to overly conservative uncertainty bounds. In this study, we use Machine Learning (ML) models to moderate the EPM perturbations and introduce our physics constrained machine learning framework for turbulence model UQ. We test this framework in multiple problems to show that it leads to improved calibration of the uncertainty estimates.

Athanasios Mamakos, Dominik Rose, Anastasios Melas et al.

The upcoming Euro 7 regulation for Heavy-Duty (HD) vehicles is calling for a further tightening of the Solid Particle Number (SPN) emissions by means of both lowering the applicable limits and shifting the lowest detectable size from 23 nm (SPN₂₃) to 10 nm (SPN₁₀). A late-technology diesel HD truck was tested on a chassis dynamometer in order to assess the necessary particle filtration requirements for a continuously regenerating system. The study showed that passive regeneration under real-world operating conditions can lead to a significant release of SPN₁₀ particles from the current technology Diesel Particulate Filter (DPF) when soot-loaded, even exceeding the currently applicable emission limits. The actual emissions during passive regeneration and following the clean-up of the DPF exceeded the proposed Euro 7 limits by more than an order of magnitude. A prototype DPF, exhibiting a 99% filtration efficiency when clean, was shown to effectively control SPN₁₀ emissions under both operating conditions. The shift to SPN₁₀ also necessitates control of nanoparticles forming inside the Selective Catalytic Reduction (SCR) system, which for the tested truck exceeded the proposed (hot) limit by up to 56%. A dedicated particle filter specifically designed to capture these particles was also evaluated, showing a better than 60% efficiency. The key message of this study is that SPN emissions can be kept at low levels under all conditions.

Pascal Krutz, André Leonhardt, Alexander Graf et al.

Given the use of high-strength steels to achieve lightweight construction goals, conventional shear-cutting processes are reaching their limits. Therefore, so-called high-speed impact cutting (HSIC) is used to achieve the required cut surface qualities. A new machine concept consisting of linear motors and an impact mass is presented to investigate HSIC. It allows all relevant parameters to be flexibly adjusted and measured. The design and construction of the test bench, as well as the mechanism for coupling the impact mass, are described. To validate the theoretically determined process speeds, the cutting process was recorded with high-speed cameras, and HSIC with a mild deep-drawing steel sheet was performed. It was discovered that very good cutting edges could be produced, which showed a significantly lower hardening depth than slowly cut reference samples. In addition, HSIC was numerically modelled in LS-DYNA, and the calculated cutting edges were compared with the real ones. With the help of adaptive meshing, a very good agreement for the cutting edges could be achieved. The results show the great potential of using a linear motor in HSIC.

Ian G. Malone, Kaleb E. Smith, Morgan E. Urdaneta et al.

Epilepsy affects millions of people, reducing quality of life and increasing risk of premature death. One-third of epilepsy cases are drug-resistant and require surgery for treatment, which necessitates localizing the seizure onset zone (SOZ) in the brain. Attempts have been made to use cortico-cortical evoked potentials (CCEPs) to improve SOZ localization but none have been successful enough for clinical adoption. Here, we compare the performance of ten machine learning classifiers in localizing SOZ from CCEP data. This preliminary study validates a novel application of machine learning, and the results establish our approach as a promising line of research that warrants further investigation. This work also serves to facilitate discussion and collaboration with fellow machine learning and/or epilepsy researchers.

Santiago Miret, Kin Long Kelvin Lee, Carmelo Gonzales et al.

We present the Open MatSci ML Toolkit: a flexible, self-contained, and scalable Python-based framework to apply deep learning models and methods on scientific data with a specific focus on materials science and the OpenCatalyst Dataset. Our toolkit provides: 1. A scalable machine learning workflow for materials science leveraging PyTorch Lightning, which enables seamless scaling across different computation capabilities (laptop, server, cluster) and hardware platforms (CPU, GPU, XPU). 2. Deep Graph Library (DGL) support for rapid graph neural network prototyping and development. By publishing and sharing this toolkit with the research community via open-source release, we hope to: 1. Lower the entry barrier for new machine learning researchers and practitioners that want to get started with the OpenCatalyst dataset, which presently comprises the largest computational materials science dataset. 2. Enable the scientific community to apply advanced machine learning tools to high-impact scientific challenges, such as modeling of materials behavior for clean energy applications. We demonstrate the capabilities of our framework by enabling three new equivariant neural network models for multiple OpenCatalyst tasks and arrive at promising results for compute scaling and model performance.