With the rapid rise of AI coding agents, the fundamental premise of what it means to be a software engineer is in question. In this vision paper, we re-examine what it means for an AI agent to be considered a software engineer and then critically think about what makes such an agent trustworthy. \textit{Grounded} in established definitions of software engineering (SE) and informed by recent research on agentic AI systems, we conceptualise AI software engineers as participants in human-AI SE teams composed of human software engineers and AI models and tools, and we distinguish trustworthiness as a key property of these systems and actors rather than a subjective human attitude. Based on historical perspectives and emerging visions, we identify key dimensions that contribute to the trustworthiness of AI software engineers, spanning technical quality, transparency and accountability, epistemic humility, and societal and ethical alignment. We further discuss how trustworthiness can be evaluated and demonstrated, highlighting a fundamental trust measurement gap: not everything that matters for trust can be easily measured. Finally, we outline implications for the design, evaluation, and governance of AI SE systems, advocating for an ethics-by-design approach to enable appropriate trust in future human-AI SE teams.
The reconstruction, management, and optimization of gas pipelines is of significant importance for solving modern engineering problems. This paper presents innovative methodologies aimed at the effective reconstruction of gas pipelines under unstable conditions. The research encompasses the application of machine learning and optimization algorithms, targeting the enhancement of system reliability and the optimization of interventions during emergencies. The findings of the study present engineering solutions aimed at addressing the challenges in real-world applications by comparing the performance of various algorithms. Consequently, this work contributes to the advancement of cutting-edge approaches in the field of engineering and opens new perspectives for future research. A highly reliable and efficient technological Figure has been proposed for managing emergency processes in gas transportation based on the principles of the reconstruction phase. For complex gas pipeline systems, new approaches have been investigated for the modernization of existing control process monitoring systems. These approaches are based on modern achievements in control theory and information technology, aiming to select emergency and technological modes. One of the pressing issues is to develop a method to minimize the transmission time of measured and controlled data on non-stationary flow parameters of gas networks to dispatcher control centers. Therefore, the reporting Figures obtained for creating a reliable information base for dispatcher centers using modern methods to efficiently manage the gas dynamic processes of non-stationary modes are of particular importance.
Segmentation of retinal vessels from fundus images is critical for diagnosing diseases such as diabetes and hypertension. However, the inherent challenges posed by the complex geometries of vessels and the highly imbalanced distribution of thick versus thin vessel pixels demand innovative solutions for robust feature extraction. In this paper, we introduce DAF-UNet, a novel architecture that integrates advanced modules to address these challenges. Specifically, our method leverages a pre-trained deformable convolution (DC) module within the encoder to dynamically adjust the sampling positions of the convolution kernel, thereby adapting the receptive field to capture irregular vessel morphologies more effectively than traditional convolutional approaches. At the network’s bottleneck, an enhanced atrous spatial pyramid pooling (ASPP) module is employed to extract and fuse rich, multi-scale contextual information, significantly improving the model’s capacity to delineate vessels of varying calibers. Furthermore, we propose a hybrid loss function that combines pixel-level and segment-level losses to robustly address the segmentation inconsistencies caused by the disparity in vessel thickness. Experimental evaluations on the DRIVE and CHASE_DB1 datasets demonstrated that DAF-UNet achieved a global accuracy of 0.9572/0.9632 and a Dice score of 0.8298/0.8227, respectively, outperforming state-of-the-art methods. These results underscore the efficacy of our approach in precisely capturing fine vascular details and complex boundaries, marking a significant advancement in retinal vessel segmentation.
Abstract The micromechanical properties of organic matter (OM) and organic pore structures in the over-mature stage are crucial for determining shale reservoir quality and assessing shale gas resource potential. However, there is still debate about the influence of micromechanical properties of OM on the micro-mesopore structures in over-mature shale. In this study, shale cores from the Niutitang Formation have been specifically chosen for OM isolation, adsorption testing, atomic force microscopy examination, and focused ion beam scanning electron microscopy (FIB-SEM) analysis to assess the micromechanical properties of OM and pore structures. The findings indicate that organic micropores and mesopores predominantly exhibit elliptical, circular, or irregular shapes. Organic pores mainly provide pore volume (PV) and specific surface area (SSA) of shale. In the over-mature stage, residual kerogen and pyrobitumen transition towards a graphite structure, increasing Young’s modulus of OM. Additionally, as thermal maturity increases, the absence of a rigid mineral framework and pore fluid pressure results in the compaction of pores, leading to a decrease in PV and SSA. The organic micropores are more vulnerable to collapse and compaction because of the increased brittleness of OM. The organic micropores and mesopores gradually evolve from regular circular and elliptical shapes to irregular shapes during the over-mature stage. The research findings provide valuable insights into the micromechanical mechanism of pore evolution in over-mature marine shale within complex structural regions.
We explore a novel approach to achieving anisotropic thermal photon tunneling, inspired by the concept of parity-time symmetry in quantum physics. Our method leverages the modulation of constitutive optical parameters, oscillating between loss and gain regimes. This modulation reveals a variety of distinct effects in thermal photon behavior and dispersion. Specifically, we identify complex tunneling modes through gain-loss engineering, which include thermal photonic defect states and Fermi-arc-like phenomena, which surpass those achievable through traditional polariton engineering. Our research also elucidates the laws governing the evolution of radiative energy in the presence of gain and loss interactions, and highlights the unexpected inefficacy of gain in enhancing thermal photon energy transport compared to systems characterized solely by loss. This study not only broadens our understanding of thermal photon tunneling but also establishes a versatile platform for manipulating photon energy transport, with potential applications in thermal management, heat science, and the development of advanced energy devices.
The rapid development of deep learning techniques, improved computational power, and the availability of vast training data have led to significant advancements in pre-trained models and large language models (LLMs). Pre-trained models based on architectures such as BERT and Transformer, as well as LLMs like ChatGPT, have demonstrated remarkable language capabilities and found applications in Software engineering. Software engineering tasks can be divided into many categories, among which generative tasks are the most concern by researchers, where pre-trained models and LLMs possess powerful language representation and contextual awareness capabilities, enabling them to leverage diverse training data and adapt to generative tasks through fine-tuning, transfer learning, and prompt engineering. These advantages make them effective tools in generative tasks and have demonstrated excellent performance. In this paper, we present a comprehensive literature review of generative tasks in SE using pre-trained models and LLMs. We accurately categorize SE generative tasks based on software engineering methodologies and summarize the advanced pre-trained models and LLMs involved, as well as the datasets and evaluation metrics used. Additionally, we identify key strengths, weaknesses, and gaps in existing approaches, and propose potential research directions. This review aims to provide researchers and practitioners with an in-depth analysis and guidance on the application of pre-trained models and LLMs in generative tasks within SE.
In this paper, a novel axial flux reversal permanent magnet (PM) (AFRPM) machine, which combines the merit of the axial flux PM (AFPM) machine and flux reversal PM machine, is proposed. Meanwhile, according to the characteristics of the machine, a method to reduce the torque ripple is presented and researched. Firstly, the topology and operation principle of the machine are analyzed. The power-size equation is derived and the design scheme is confirmed to obtain the structure parameters of the machine. Then, the general electromagnetic performances are analyzed. Secondly, the magnetomotive force (MMF)-permeance model is established. The air gap flux density harmonic distribution principle is researched from the air gap field modulation perspective. The torque generation mechanism is studied by analyzing the influences of the armature reaction air gap flux density harmonics and no-load air gap flux density harmonics on the electromagnetic torque. Furthermore, in order to reduce the torque ripple, an optimization method using the cos-type PMs, which suppress the harmonic pole pairs that do not contribute to the torque output, is investigated. To simplify the manufacturing process, the segmented sector PMs with different heights and arcs are employed instead of the cos-type PM. The effects of the different numbers of segmented sector PMs on the torque ripple are studied. Finally, the results indicate that the proposed machine exhibits large torque capability and high power/torque density. Meanwhile, the torque ripple is significantly reduced by the optimization method.
Registered reports are scientific publications which begin the publication process by first having the detailed research protocol, including key research questions, reviewed and approved by peers. Subsequent analysis and results are published with minimal additional review, even if there was no clear support for the underlying hypothesis, as long as the approved protocol is followed. Registered reports can prevent several questionable research practices and give early feedback on research designs. In software engineering research, registered reports were first introduced in the International Conference on Mining Software Repositories (MSR) in 2020. They are now established in three conferences and two pre-eminent journals, including Empirical Software Engineering. We explain the motivation for registered reports, outline the way they have been implemented in software engineering, and outline some ongoing challenges for addressing high quality software engineering research.
Malikasab Bagawan, Suresh T. Dundur, Rajesh Gurani
et al.
AbstractElastomers are the class of materials that are widely used in a variety of industrial, commercial, and consumer applications due to their unique mechanical properties, including high elasticity, high flexibility, and high resilience. However, there are many challenges in machining of elastomers such as poor surface finish, build up of heat, degradation of elastomers, etc. To overcome these challenges, cryogenic cooling assistance has been introduced as a means of improving the machinability of elastomers. This paper presents a soft computing approach for optimizing the surface roughness and cutting force during turning of elastomers under different lubrication conditions. Three types of elastomers, namely Nitrile Rubber (NBR), Polyurethane Rubber (PU), and Neoprene Rubber (CR), are studied, and a cryogenic fluid delivery system is employed to improve the machining process. Taguchi’s L27 array is used to vary the input parameters, and a Back-Propagation Artificial Neural Network (BPANN) model is developed to predict the cutting force and surface roughness. The cutting force and surface roughness are analyzed under different cooling conditions, cutting speeds, feeds, and depths of cut for various elastomers. The results show that changes in cutting conditions significantly affect the cutting force and that the type of lubrication used affects the cutting force by altering the material’s physical properties. Cutting force is significantly influenced by cutting conditions, and NBR requires the highest cutting force compared to PU and CR. Further, at a cutting speed of 55 m/min, a feed of 0.11 mm/rev, and a depth of cut of 0.25 mm, the cutting force for NBR (85.1 N), while for PU (75.1 N) and CR (80.3 N), respectively. Finally, with LN2 lubrication conditions, the Cutting Force decreased by 45% and Surface Roughness decreased by 16.9%. This study provides insights into the factors affecting the elastomer machining process, which can be useful for optimizing the machining process parameters and improving machining efficiency.
An asymmetric image encryption algorithm based on integer wavelet transformation (IWT) and Rivest-Shamir-Adleman (RSA) algorithm is proposed. Firstly, two plain characteristic parameters (PCP) of the plain image are extracted and two random numbers are chosen. Then, a new parameter transformation model (PTM) is constructed to do nonlinear processing for them, and three cipher characteristic parameters (CCP) are got. After applying RSA operation for CCP (seen as plain messages), three cipher messages are obtained. Secondly, a new initial value obtaining model (IOM) for all plain messages and cipher messages is established, by which initial values of 3D chaotic system are produced. Then, three chaotic sequences can be generated. Thirdly, chaotic sequences are used to confuse the plain image by a way of row-column cycle. Then, IWT operation is carried out and the above chaotic sequences are employed to confuse again the wavelet coefficients. Thereafter, inverse IWT is applied to get the confused image, realizing double confusion operations on both spatial domain and frequency domain. Finally, the confused image is diffused as a whole to get the cipher image. Experiment results explain that the proposed algorithm can realize the encryption in short time, and resist effectively against brute-force attack and noise attack.
In the last decade, Single-Board Computers (SBCs) have been employed more frequently in engineering and computer science both to technical and educational levels. Several factors such as the versatility, the low-cost, and the possibility to enhance the learning process through technology have contributed to the educators and students usually employ these devices. However, the implications, possibilities, and constraints of these devices in engineering and Computer Science (CS) education have not been explored in detail. In this systematic literature review, we explore how the SBCs are employed in engineering and computer science and what educational results are derived from their usage in the period 2010-2020 at tertiary education. For that, 154 studies were selected out of n=605 collected from the academic databases Ei Compendex, ERIC, and Inspec. The analysis was carried-out in two phases, identifying, e.g., areas of application, learning outcomes, and students and researchers' perceptions. The results mainly indicate the following aspects: (1) The areas of laboratories and e-learning, computing education, robotics, Internet of Things (IoT), and persons with disabilities gather the studies in the review. (2) Researchers highlight the importance of the SBCs to transform the curricula in engineering and CS for the students to learn complex topics through experimentation in hands-on activities. (3) The typical cognitive learning outcomes reported by the authors are the improvement of the students' grades and the technical skills regarding the topics in the courses. Concerning the affective learning outcomes, the increase of interest, motivation, and engagement are commonly reported by the authors.
Chenyang Yang, Rachel Brower-Sinning, Grace A. Lewis
et al.
In spite of machine learning's rapid growth, its engineering support is scattered in many forms, and tends to favor certain engineering stages, stakeholders, and evaluation preferences. We envision a capability-based framework, which uses fine-grained specifications for ML model behaviors to unite existing efforts towards better ML engineering. We use concrete scenarios (model design, debugging, and maintenance) to articulate capabilities' broad applications across various different dimensions, and their impact on building safer, more generalizable and more trustworthy models that reflect human needs. Through preliminary experiments, we show capabilities' potential for reflecting model generalizability, which can provide guidance for ML engineering process. We discuss challenges and opportunities for capabilities' integration into ML engineering.
There is a growing interest in developing an advanced technique to manufacture titanium alloy for larger holes in the aviation and marine industries. While the application of larger holes is limited due to the low machinability and high cost. The deep-hole trepanning is investigated to manufacture the titanium alloy in a high-efficient low-cost manner, and the technical challenges during the deep-hole trepanning process are studied in this article. Designs of the deep-hole trepanning tool and experiments under different cutting speeds and feed are carried out. Chip morphology, axial cutting force, and tool wear are analyzed. The optimal process parameters are achieved. This study provides a practical solution to process titanium alloy for potential industrial applications.
Two-dimensional antiferromagnetic semiconductors have triggered significant attention due to their unique physical properties and broad application. Based on first-principles calculations, a novel two-dimensional (2D) antiferromagnetic material MnSi2N4 monolayer is predicted. The calculation results show that the two-dimensional MnSi2N4 prefers an antiferromagnetic state with a small band gap of 0.26 eV. MnSi2N4 has strong antiferromagnetic coupling which can be effectively tuned under strain. Interestingly, the MnSi2N4 monolayer exhibits a half-metallic ferromagnetic properties under an external magnetic field, in which the spin-up electronic state displays a metallic property, while the spin-down electronic state exhibits a semiconducting characteristic. Therefore, 100% spin polarization can be achieved. Two-dimensional MnSi2N4 monolayer has potential application in the field of high-density information storage and spintronic devices.
Concetta Di Natale, Concetta Di Natale, Concetta Di Natale
et al.
Bacterial cellulose (BC) is a highly pure form of cellulose produced by bacteria, which possesses numerous advantages such as good mechanical properties, high chemical flexibility, and the ability to assemble in nanostructures. Thanks to these features, it achieved a key role in the biomedical field and in drug delivery applications. BC showed its ability to modulate the release of several drugs and biomolecules to the skin, thus improving their clinical outcomes. This work displays the loading of a 3D BC nanonetwork with an innovative drug delivery nanoemulsion system. BC was optimized by static culture of SCOBY (symbiotic colony of bacteria and yeast) and characterized by morphological and ultrastructural analyses, which indicate a cellulose fiber diameter range of 30–50 nm. BC layers were then incubated at different time points with a nanocarrier based on a secondary nanoemulsion (SNE) previously loaded with a well-known antioxidant and anti-inflammatory agent, namely, coenzyme-Q10 (Co-Q10). Incubation of Co-Q10–SNE in the BC nanonetwork and its release were analyzed by fluorescence spectroscopy.
Rachel L. Mugge, Jennifer L. Salerno, Leila J. Hamdan
Marine biofilms are essential biological components that transform built structures into artificial reefs. Anthropogenic contaminants released into the marine environment, such as crude oil and chemical dispersant from an oil spill, may disrupt the diversity and function of these foundational biofilms. To investigate the response of marine biofilm microbiomes from distinct environments to contaminants and to address microbial functional response, biofilm metagenomes were analyzed from two short-term microcosms, one using surface seawater (SSW) and the other using deep seawater (DSW). Following exposure to crude oil, chemical dispersant, and dispersed oil, taxonomically distinct communities were observed between microcosms from different source water challenged with the same contaminants and higher Shannon diversity was observed in SSW metagenomes. Marinobacter, Colwellia, Marinomonas, and Pseudoalteromonas phylotypes contributed to driving community differences between SSW and DSW. SSW metagenomes were dominated by Rhodobacteraceae, known biofilm-formers, and DSW metagenomes had the highest abundance of Marinobacter, associated with hydrocarbon degradation and biofilm formation. Association of source water metadata with treatment groups revealed that control biofilms (no contaminant) harbor the highest percentage of significant KEGG orthologs (KOs). While 70% functional similarity was observed among all metagenomes from both experiments, functional differences between SSW and DSW metagenomes were driven primarily by membrane transport KOs, while functional similarities were attributed to translation and signaling and cellular process KOs. Oil and dispersant metagenomes were 90% similar to each other in their respective experiments, which provides evidence of functional redundancy in these microbiomes. When interrogating microbial functional redundancy, it is crucial to consider how composition and function evolve in tandem when assessing functional responses to changing environmental conditions within marine biofilms. This study may have implications for future oil spill mitigation strategies at the surface and at depth and also provides information about the microbiome functional responses of biofilms on steel structures in the marine built environment.
Julianna C. da Silva, Luis F. F. Costa, Daniella P. dos Santos
et al.
ABSTRACT This study aimed to carry out an economical and productive analysis of carrot production using different irrigation depths and doses of fertilizer. A randomized block design was used arranged in a 6 × 4 factorial scheme, with three replicates. The treatments were constituted by six irrigation depths: (L1: 210.5, L2: 315.7, L3: 421.0, L4: 526.2, L5: 631.5, and L6: 736.7 mm) and four doses of fertilizer: (F1: 226.9, F2: 340.3, F3: 453.8 and F4: 567.2 kg ha-1) applied via fertigation. At the end of the cycle, four carrot roots were collected per plot to estimate yield. The maximum estimated yield of the carrot was 95.4 t ha-1, obtained using 478.1 mm of water and 538.8 kg ha-1 of fertilizer. The best economic return was achieved with 482.0 mm of water and 460.0 kg ha-1 of fertilizer, giving 95.0 t ha-1. The combination of irrigation and fertilizer allows lesser amount of both to be used, giving greater response than when applied separately.
Unlabeled sensing is a linear inverse problem where the measurements are scrambled under an unknown permutation leading to loss of correspondence between the measurements and the rows of the sensing matrix. Motivated by practical tasks such as mobile sensor networks, target tracking and the pose and correspondence estimation between point clouds, we study a special case of this problem restricting the class of permutations to be local and allowing for multiple views. In this setting, namely multi-view unlabeled sensing under local permutations, previous results and algorithms are not directly applicable. In this paper, we propose a computationally efficient algorithm, R-local unlabeled sensing (RLUS), that creatively exploits the machinery of indefinite relaxations of the graph matching problem to estimate the local permutations. Simulation results on synthetic data sets indicate that the proposed algorithm is scalable and applicable to the challenging regimes of low to moderate SNR.
Model-based systems engineering (MBSE) provides an important capability for managing the complexities of system development. MBSE empowers the formalisms of system architectures for supporting model-based requirement elicitation, specification, design, development, testing, fielding, etc. However, the modeling languages and techniques are quite heterogeneous, even within the same enterprise system, which creates difficulties for data interoperability. The discrepancies among data structures and language syntaxes make information exchange among MBSE models even more difficult, resulting in considerable information deviations when connecting data flows across the enterprise. For this reason, this paper presents an ontology based upon graphs, objects, points, properties, roles, and relationships with entensions (GOPPRRE), providing meta models that support the various lifecycle stages of MBSE formalisms. In particular, knowledge-graph models are developed to support unified model representations to further implement ontological data integration based on GOPPRRE throughout the entire lifecycle. The applicability of the MBSE formalism is verified using quantitative and qualitative approaches. Moreover, the GOPPRRE ontologies are generated from the MBSE language formalisms in a domain-specific modeling tool, \textit{MetaGraph} in order to evaluate its availiablity. The results demonstrate that the proposed ontology supports both formal structures and the descriptive logic of the systems engineering lifecycle.