Hasil untuk "Machine design and drawing"

Julian Borowski, Hinrich Emsmann, Jannis Kneule et al.

Increasing product complexity, shorter development cycles and cross-domain integration demands pose significant challenges for modern race car engineering teams. In Formula Student teams, heterogeneous toolchains, manual data exchange, late system integration, and high personnel turnover hinder efficient collaborative development and lead to repeated knowledge loss. This paper presents an integrated digital-engineering framework combining graph-based design languages (GBDL), model-to-text transformations, natural-language interactions via Large Language Models (LLMs), and Git-based version control to address these issues. By formalizing design knowledge and storing it in a centralized design graph, the framework ensures digital consistency of data and models, supports automated vehicle design variant generation, and enables seamless cross-domain integration. Through case studies of three Formula Student teams, the methodology demonstrates quantifiable reductions in design iteration time, enabling the evaluation of more than 10⁴ suspension variants within days instead of a few dozen manually created variants, while reducing hands-on engineering effort from minutes per variant to a largely unattended optimization process. The results indicate that the approach not only enhances efficiency and collaboration but also preserves design knowledge for long-term knowledge management and reuse. Looking forward, this methodology provides a scalable route toward further engineering automation, systematic variant-driven development, and early-stage design optimization supported by design languages and integrated downstream toolchains.

Farrukh A. Chishtie, John J. Clague

The Pacific Northwest faces significant seismic hazards from both great subduction earthquakes and more frequent in-slab events within the Juan de Fuca plate system. This paper presents a breakthrough shift in earthquake risk assessment by integrating geological knowledge from the natural sciences with Actor-Network Theory (ANT) and mobilities research from the social sciences to reconceptualize seismic risk through the lens of transectional networks involving human and non-human actors. We examine the translation processes through which seismic monitoring systems, building codes, emergency response protocols, geological formations, and emerging artificial intelligence/machine learning technologies co-constitute earthquake risk in the region. Drawing from recent advances in uncertainty quantification and economic impact assessment methodologies developed for climate litigation, we argue for more sophisticated measurement protocols that can capture the relational dynamics and cascading effects within seismic networks. The historical record of in-slab earthquakes, including the 24-year gap since the last magnitude 6+ event in 2001, illustrates how temporal patterns emerge from complex interactions between geological agencies and human systems. We develop a novel five-phase integrated transectional risk assessment methodology that holistically accounts for both human and non-human vulnerabilities as they emerge from dynamic network relationships across spatial, temporal, and organizational scales. This methodology operationalizes network mapping, translation analysis, transectional vulnerability assessment, integrated uncertainty quantification, and adaptive intervention design to move beyond traditional hazard-exposure-vulnerability frameworks. The transectional perspective reveals opportunities for earthquake risk reduction that go beyond traditional engineering approaches to encompass network reconfigurations, AI-enhanced monitoring systems, innovative financing mechanisms, and enhanced adaptive capacities across human-non-human assemblages. This interdisciplinary approach provides concrete pathways for developing more effective and equitable earthquake risk management strategies that recognize the agency of both geological processes and technological systems in shaping seismic resilience.

Dumpala Gangi Reddy, Ramarathnam Krishna Kumar

Three-wheeled vehicles are increasingly adopted as sustainable transport solutions, but their asymmetric design and lightweight structure make them vulnerable to ride discomfort and rollover instability. This study develops a high-fidelity 12-degrees-of-freedom (DOF) dynamic model in MATLAB/Simulink and MSC ADAMS to analyze and improve ride comfort, handling, and roll stability. The model captures longitudinal, lateral, vertical, roll, pitch, and yaw motions, along with tire dynamics represented through the Magic Formula, and is validated using real-world data from an instrumented test vehicle. In this research, both active and passive control strategies were separately implemented and studied. The active strategy involves an Active Vehicle Roll Dynamics Control (VRDC) system with an active rear suspension to suppress roll and yaw during aggressive maneuvers. The passive strategy focuses on improving rollover resistance by modulating throttle input based on sensor data from gyroscopes, accelerometers, and compasses. Simulation and experimental results show that each strategy, when applied independently, enhances roll stability, reduces yaw rate deviations, and improves handling performance. These findings demonstrate the effectiveness of both approaches in improving the safety and dynamic behavior of electric three-wheeled vehicles under real-world conditions.

Betty Mayeku, Sandra Hummel, Parisa Memarmoshrefi

Machine Unlearning (MU) has emerged as a promising approach to addressing persistent challenges in Machine Learning (ML) systems. By enabling the selective removal of learned data, MU introduces protective, corrective, and adaptive capabilities that are central to advancing Responsible and Adaptive AI. However, despite its growing prominence in other domains, MU remains underexplored within education, a sector uniquely characterized by sensitive learner data, dynamic environments, and the high-stakes implications of algorithmic decision-making. This paper examines the potential of MU as both a mechanism for operationalizing Responsible AI principles and a foundation for Adaptive AI in ML-driven educational systems. Drawing on a structured review of 42 peer-reviewed studies, the paper analyzes key MU mechanisms and technical variants, and how they contribute to the practical realization of Responsible and Adaptive AI. Four core intervention domains where MU demonstrates significant promise are identified: privacy protection, resilience to adversarial or corrupted data, fairness through bias mitigation, and adaptability to evolving contexts. Furthermore, MU interventions are mapped to the technical, ethical, and pedagogical challenges inherent in educational AI. This mapping illustrates the role of MU as a strategic mechanism for enhancing compliance, reinforcing ethical safeguards, and supporting adaptability by ensuring that models remain flexible, maintainable, and contextually relevant over time. As a conceptual contribution, the paper introduces MU4RAAI, a reference architecture integrating MU within Responsible and Adaptive AI frameworks for educational contexts. MU is thus positioned not merely as a data deletion process but as a transformative approach for ensuring that educational AI systems remain ethical, adaptive, and trustworthy.

Lirandë Pira, Airin Antony, Nayanthara Prathap et al.

Photonic chip design has seen significant advancements with the adoption of inverse design methodologies, offering flexibility and efficiency in optimizing device performance. However, the black-box nature of the optimization approaches, such as those used in inverse design in order to minimize a loss function or maximize coupling efficiency, poses challenges in understanding the outputs. This challenge is prevalent in machine learning-based optimization methods, which can suffer from the same lack of transparency. To this end, interpretability techniques address the opacity of optimization models. In this work, we apply interpretability techniques from machine learning, with the aim of gaining understanding of inverse design optimization used in designing photonic components, specifically two-mode multiplexers. We base our methodology on the widespread interpretability technique known as local interpretable model-agnostic explanations, or LIME. As a result, LIME-informed insights point us to more effective initial conditions, directly improving device performance. This demonstrates that interpretability methods can do more than explain models -- they can actively guide and enhance the inverse-designed photonic components. Our results demonstrate the ability of interpretable techniques to reveal underlying patterns in the inverse design process, leading to the development of better-performing components.

Pierre-Marie Chauvin, Angèle Merlin, Xavier Fresquet et al.

This paper explores the integration of generative AI into the fashion design process. Drawing on insights from the January 2025 seminar ``Tisser le futur,'' it investigates how AI reshapes creative workflows, from ideation to prototyping, while interrogating the ethical, aesthetic, and labor implications. The paper highlights co-creative dynamics between humans and machines, the potential for aesthetic innovation, and the environmental and cultural challenges of algorithmic design.

Yeong-Maw Hwang, Tsung-Han Ho, Yung-Fa Huang et al.

Deep drawing has been practiced in various manufacturing industries for many years. With the aid of stamping equipment, materials are sheared to different shapes and dimensions for users. Meanwhile, through artificial intelligence (AI) training, machines can make decisions or perform various functions. The aim of this study is to discuss the geometric and process parameters for A7075 in deep drawing and derive the formable regions of sound products for different forming parameters. Four parameters—forming temperature, punch speed, blank diameter and thickness—are used to investigate their effects on the forming results. Through finite element simulation, a database is established and used for machine learning (ML) training and validation to derive an AI prediction model. Importing the forming parameters into this prediction model can obtain the forming results rapidly. To validate the formable regions of sound products, several experiments are conducted and the results are compared with the prediction results to verify the feasibility of applying ML to deep drawing processes of aluminum alloy A7075 and the reliability of the AI prediction model.

Anna Granà, Salvatore Curto, Andrea Petralia et al.

The introduction of connected automated vehicles (CAVs) on freeways raises significant challenges, particularly in interactions with human-driven vehicles, impacting traffic flow and safety. This study employs traffic microsimulation and surrogate safety assessment measures software to delve into CAV–human driver interactions, estimating potential conflicts. While previous research acknowledges that human drivers adjust their behavior when sharing the road with CAVs, the underlying reasons and the extent of associated risks are not fully understood yet. The study focuses on how CAV presence can diminish conflicts, employing surrogate safety measures and real-world mixed traffic data, and assesses the safety and performance of freeway interchange configurations in Italy and the US across diverse urban contexts. This research proposes tools for optimizing urban layouts to minimize conflicts in mixed traffic environments. Results reveal that adding auxiliary lanes enhances safety, particularly for CAVs and rear-end collisions. Along interchange ramps, an exclusive CAV stream performs similarly to human-driven ones in terms of longitudinal conflicts, but mixed traffic flows, consisting of both CAVs and human-driven vehicles, may result in more conflicts. Notably, when CAVs follow human-driven vehicles in near-identical conditions, more conflicts arise, emphasizing the complexity of CAV integration and the need for careful safety measures and roadway design considerations.

Erőss László Dániel, Markovits Tamás

Additive manufacturing and topology optimization provide new possibilities to produce complex parts. They can be used separately but with joint applications as a mutually reinforcing solution in component development tasks. The results obtained using the design software can be refined even further depending on the specific goal set. This paper deals with mass reduction with stiffness-based topology optimization of a structural component. The effect of different design spaces, load cases, and design parameters were examined. Then, the new part was validated with FEA simulation. After the validation, the part was prepared for 3D metal printing. Based on the research results, we present a methodology that can be used as a solution considering the software’s limitations and the development of the specific component. Applying the methodology developed in the research makes it possible to achieve mass minimization on other parts with a similar method.

Saltanat Nurkusheva, Michał Bembenek, Maciej Berdychowski et al.

This article presents a case study on estimating the real service inspection intervals for German-brand passenger cars in Kazakhstan and Poland. This study aimed to identify disparities between the official recommendations of manufacturers for car maintenance and the real data collected in these two countries. The following passenger cars were examined: Audi A6, Q5, and Q8; Porsche Cayenne and Cayenne coupe; and Volkswagen Passat, Polo, Teramont, Tiguan, Touareg, Arteon, Golf, T-Cross, Tiguan all space, Touran, T-Roc, and Up. To assess the difference between real and recommended values, the manufacturer criteria of a recommended mileage of 15,000 and 30,000 km or a time frame of 365 and 730 days to the first service inspection were applied. The data analysis showed that in Kazakhstan, 31.4% of cars did not meet the warranty conditions, while in Poland, it was 21.0%. The dominant criterion that was not met was the time criterion. The assessment of these factors emphasizes the importance of customizing vehicle maintenance schedules to the specific conditions and driving behaviors prevalent in each country. The practical contribution of the article lies in uncovering the discrepancies between official manufacturer recommendations for car maintenance and the actual data collected in Kazakhstan and Poland. By identifying specific models, Volkswagen Touareg and Tiguan in Kazakhstan and Volkswagen Up in Poland, for which the maintenance intervals deviated significantly from those recommended, this study offers valuable insights for optimizing service schedules and improving the efficiency of maintenance practices in these countries. From a scientific perspective, this article contributes by providing empirical evidence of real-world maintenance behaviors for German-brand passenger cars.

Yiwei Wang, Tao Yang, Hailin Huang et al.

This paper presents a data-driven electrical machine design (EMD) framework using wound-rotor synchronous generator (WRSG) as a design example. Unlike traditional preliminary EMD processes that heavily rely on expertise, this framework leverages an artificial-intelligence based expert database, to provide preliminary designs directly from user specifications. Initial data is generated using 2D finite element (FE) machine models by sweeping fundamental design variables including machine length and diameter, enabling scalable machine geometry with machine performance for each design is recorded. This data trains a Metamodel of Optimal Prognosis (MOP)-based surrogate model, which maps design variables to key performance indicators (KPIs). Once trained, guided by metaheuristic algorithms, the surrogate model can generate thousands of geometric scalable designs, covering a wide power range, forming an AI expert database to guide future preliminary design. The framework is validated with a 30kVA WRSG design case. A prebuilt WRSG database, covering power from 10 to 60kVA, is validated by FE simulation. Design No.1138 is selected from database and compared with conventional design. Results show No.1138 achieves a higher power density of 2.21 kVA/kg in just 5 seconds, compared to 2.02 kVA/kg obtained using traditional method, which take several days. The developed AI expert database also serves as a high-quality data source for further developing AI models for automatic electrical machine design.

Eduardo Tomanik, Antonio J. Jimenez-Reyes, Victor Tomanik et al.

Transient car emission tests generate huge amount of test data, but their results are usually evaluated only using their “accumulated” cycle values according to the homologation limits. In this work, two machine learning models were developed and applied to a truck RDE test and two light-duty vehicle chassis emission tests. Different from the conventional approach, the engine parameters and fuel consumption were acquired from the Engine Control Unit, not from the test measurement equipment. Instantaneous engine values were used as input in machine-learning-based digital twins. This novel approach allows for much less costly vehicle tests and optimizations. The paper’s novel approach and developed digital twins model were able to predict both instantaneous and accumulated fuel consumption with good accuracy, and also for tests cycles different to the one used to train the model.

Achim Kampker, Heiner Heimes, Benjamin Dorn et al.

In the European Union (EU), road transport contributes a major proportion of the total greenhouse gas (GHG) emissions, of which a significant amount is caused by heavy-duty commercial vehicles (CV). The increasing number of emission regulations and penalties by the EU have forced commercial vehicle manufacturers to investigate powertrain technologies other than conventional internal combustion engines (ICE). Since vehicle economics plays an important role in purchase decisions and the powertrain of a battery electric vehicle (BEV) contributes to about 8–20% of the total vehicle cost and the electric machine (EM) alone contributes to 33–43% of the drivetrain cost, it is necessary to analyze suitable EM topologies for the powertrain. In this paper, the authors aim to analyze the technical and cost aspects of an EM for electric commercial vehicles (ECV). Based on prior research and literature on this subject, an appropriate methodology for selecting suitable geometrical parameters of an e-machine for the use case of a heavy-duty vehicle is developed using MATLAB and Simulink tools. Then, for the economic analysis of the e-machine, reference ones are used, and their design parameters and cost structures are utilized to develop a cost function. Different use cases are evaluated according to the vehicle’s application. The results for a use case are compared by varying the design parameters to find the most cost-effective EM. Later, an analysis is performed on other decisive factors for EM selection. This highlights the importance of collaborative consideration of technological as well as the economic aspects of EMs for different use cases in ECVs. The method developed in this work contributes to understand the economic aspect of EMs as well as considering their performance factors. State-of-the-art methods and research are used to develop a novel methodology that helps with the selection of the initial geometry of the electric motor during the design process, which can serve to aid future designers and converters of electric heavy-duty vehicles.

Teaya Yang, David Hathcock, Yuchao Chen et al.

We propose a novel design paradigm for multistate machines where transitions from one state to another are organized by bifurcations of multiple equilibria of the energy landscape describing the collective interactions of the machine components. This design paradigm is attractive since, near bifurcations, small variations in a few control parameters can result in large changes to the system's state providing an emergent lever mechanism. Further, the topological configuration of transitions between states near such bifurcations ensures robust operation, making the machine less sensitive to fabrication errors and noise. To design such machines, we develop and implement a new efficient algorithm that searches for interactions between the machine components that give rise to energy landscapes with these bifurcation structures. We demonstrate a proof of concept for this approach by designing magneto elastic machines whose motions are primarily guided by their magnetic energy landscapes and show that by operating near bifurcations we can achieve multiple transition pathways between states. This proof of concept demonstration illustrates the power of this approach, which could be especially useful for soft robotics and at the microscale where typical macroscale designs are difficult to implement.

Saeed Khaki, Akhouri Abhinav Aditya, Zohar Karnin et al.

Drift in machine learning refers to the phenomenon where the statistical properties of data or context, in which the model operates, change over time leading to a decrease in its performance. Therefore, maintaining a constant monitoring process for machine learning model performance is crucial in order to proactively prevent any potential performance regression. However, supervised drift detection methods require human annotation and consequently lead to a longer time to detect and mitigate the drift. In our proposed unsupervised drift detection method, we follow a two step process. Our first step involves encoding a sample of production data as the target distribution, and the model training data as the reference distribution. In the second step, we employ a kernel-based statistical test that utilizes the maximum mean discrepancy (MMD) distance metric to compare the reference and target distributions and estimate any potential drift. Our method also identifies the subset of production data that is the root cause of the drift. The models retrained using these identified high drift samples show improved performance on online customer experience quality metrics.

Maxwell T. West, Shu-Lok Tsang, Jia S. Low et al.

Machine learning algorithms are powerful tools for data driven tasks such as image classification and feature detection, however their vulnerability to adversarial examples - input samples manipulated to fool the algorithm - remains a serious challenge. The integration of machine learning with quantum computing has the potential to yield tools offering not only better accuracy and computational efficiency, but also superior robustness against adversarial attacks. Indeed, recent work has employed quantum mechanical phenomena to defend against adversarial attacks, spurring the rapid development of the field of quantum adversarial machine learning (QAML) and potentially yielding a new source of quantum advantage. Despite promising early results, there remain challenges towards building robust real-world QAML tools. In this review we discuss recent progress in QAML and identify key challenges. We also suggest future research directions which could determine the route to practicality for QAML approaches as quantum computing hardware scales up and noise levels are reduced.

Binyang Song, Rui Zhou, Faez Ahmed

In the rapidly advancing field of multi-modal machine learning (MMML), the convergence of multiple data modalities has the potential to reshape various applications. This paper presents a comprehensive overview of the current state, advancements, and challenges of MMML within the sphere of engineering design. The review begins with a deep dive into five fundamental concepts of MMML:multi-modal information representation, fusion, alignment, translation, and co-learning. Following this, we explore the cutting-edge applications of MMML, placing a particular emphasis on tasks pertinent to engineering design, such as cross-modal synthesis, multi-modal prediction, and cross-modal information retrieval. Through this comprehensive overview, we highlight the inherent challenges in adopting MMML in engineering design, and proffer potential directions for future research. To spur on the continued evolution of MMML in engineering design, we advocate for concentrated efforts to construct extensive multi-modal design datasets, develop effective data-driven MMML techniques tailored to design applications, and enhance the scalability and interpretability of MMML models. MMML models, as the next generation of intelligent design tools, hold a promising future to impact how products are designed.

Peter Salter

AbstractArt, architecture and urbanism practice Metis was founded by Professor Mark Dorrian, Forbes Chair in Architecture, and Adrian Hawker, MArch Programme Director, at the University of Edinburgh. Metis has a long history of proposing extraordinary architectures, whether in building form or as interventions in existing spaces. Architect and academic Peter Salter uses the lens of Francis Bacon's painting to view their creative tactics.

Jinjin Rong, Xu Ji, Xin Fang et al.

Due to the rapid growth of the elderly around the world, the artificial intelligence control framework can collect information and apply it and perform other tasks. Artificial intelligence plays an important role in focusing on the elderly. For example, it can improve the relationship between the elderly and family members or nursing teams. In addition, AI chat robot can communicate with the elderly without obstacles and can remind the elderly when to take medicine, regular physical examination, etc. A significant number of the AI applications on cell phones accessible today could screen wellbeing information, like every day exercises, diet, and surprisingly the senior’s way of life, in a less nosy way. In such cases, it could help in expecting and, subsequently, forestalling any conceivable hypertension or unpredictable heart rate. Essentially, mechanical ‘pets’ are likewise assisting with fighting off feelings of loneliness, while additionally assisting with upgrading patient consideration simultaneously. One model is Tombot, a little dog like model, which was made to diminish misery and tension among dementia patients. Its head developments, looks, and swaying tail feel basically the same as the real thing, causing occupants to feel as though they have their very own pet to really focus on. One of the issues that growing societies are presently facing is the care of elderly individuals. The dearth of skilled workers in the senior healthcare setting has been exacerbated by the worldwide shift of aging populations. There might be an enhanced need for old nursing since the global older demographic is expected to nearly triple in the coming three decades. There are advancements in computer technologies for supporting the aged plus associated caregivers, checking their wellbeing, and offering company to them. Given the global elderly demographic development possibilities, it is no coincidence that the aided care market is drawing fast advancement, rendering health management for nurses a breeze. While the world’s governments manage the aging population next years, these ideas will become extremely vital. They will almost certainly encounter economic and political demands to modify state medical care management, retirement benefits, and social security in order to meet the needs of an aging population. Considering that the demand for physicians is growing, a necessity has developed to deliver individualized care for the aged as well as to respond appropriately in emergencies. As a result, in the technological society, healthcare is exploring artificial intelligence to deliver personalized treatment to individuals in need. The challenges of the aged are determined in this study, and answers are supplied via a tailored computer (robot). With the crucial details given via Internet of Things gadgets, emergency events may be foreseen relatively promptly, and appropriate actions can be proposed by AI technology. Individuals’ vital health information is collected using the Internet of Things based on smart technology. The information is evaluated, and decisions are made by AI, while the developed machine performs the appropriate task. This research, therefore, looks at the material design of medical products for elderly people based on artificial intelligence. It goes further and explains some of the challenges encountered in the process and possible remedies.

Meng Hua, Qingqing Wu, Wen Chen et al.

This paper studies a communication-centric integrated sensing and communication (ISAC) system, where a multi-antenna base station (BS) simultaneously performs downlink communication and target detection. A novel target detection and information transmission protocol is proposed, where the BS executes the channel estimation and beamforming successively and meanwhile jointly exploits the pilot sequences in the channel estimation stage and user information in the transmission stage to assist target detection. We investigate the joint design of pilot matrix, training duration, and transmit beamforming to maximize the probability of target detection, subject to the minimum achievable rate required by the user. However, designing the optimal pilot matrix is rather challenging since there is no closed-form expression of the detection probability with respect to the pilot matrix. To tackle this difficulty, we resort to designing the pilot matrix based on the information-theoretic criterion to maximize the mutual information (MI) between the received observations and BS-target channel coefficients for target detection. We first derive the optimal pilot matrix for both channel estimation and target detection, and then propose an unified pilot matrix structure to balance minimizing the channel estimation error (MSE) and maximizing MI. Based on the proposed structure, a low-complexity successive refinement algorithm is proposed. Simulation results demonstrate that the proposed pilot matrix structure can well balance the MSE-MI and the Rate-MI tradeoffs, and show the significant region improvement of our proposed design as compared to other benchmark schemes. Furthermore, it is unveiled that as the communication channel is more correlated, the Rate-MI region can be further enlarged.