The research team, led by Norashikin Ahmad Kamal of Universiti Teknologi MARA in Shah Alam, focused on a course called Sustainable Management, taught within a Master of Civil Engineering program at a Malaysian public university. Civil engineers occupy a peculiar position in the sustainability transition: they design the bridges, water systems, and urban infrastructure that either lock societies into carbon-intensive patterns or unlock resilient, low-impact futures. If sustainability principles fail to take root in civil engineering education, the consequences are literally built into the landscape for generations. That stakes-laden reality is precisely why the researchers chose to dissect this course with unusual forensic intensity.

Methodologically, the study is a tour de force of educational analytics. The team employed a mixed-methods design anchored in Continuous Quality Improvement, or CQI, a discipline borrowed from industrial engineering that treats education as a production process subject to the same statistical scrutiny as a manufacturing line. Quantitative data came from raw score marks across ten semesters, supplemented by a decade’s worth of institutional CQI reports documenting how the course was adjusted, reviewed, and re-adjusted over time. These numbers were then corroborated against qualitative insights harvested from Student Feedback Online, the university’s systematic feedback platform, allowing the researchers to triangulate what students achieved with what students experienced.

The results split cleanly into two categories: competencies that thrive and competencies that wobble. Communication, professionalism, and collaboration demonstrated consistently robust performance across the entire ten-semester window. Students emerged from the course able to articulate sustainability trade-offs, work in multidisciplinary teams, and conduct themselves with professional maturity. This is no small achievement, and it aligns with what accreditation bodies such as the Washington Accord increasingly demand from engineering graduates. The message from the data is that structured, well-designed coursework can reliably build the interpersonal and ethical dimensions of sustainability competence.

But the picture darkens considerably when the analysis turns to lifelong learning and problem-solving. These two competencies exhibited what the researchers describe as considerable heterogeneity and intermittent deterioration, statistical language for a troubling pattern: some semesters produced strong results, others produced marked declines, and the variance between individual students was substantial. In practical terms, the course was succeeding at teaching students what to think about sustainability while leaving unevenly developed their ability to keep learning about it independently after graduation, and to solve novel problems that no lecture has anticipated. For a field where the science, regulations, and technologies shift annually, that gap is not a footnote. It is the central vulnerability.

The heterogeneity finding deserves particular attention because it exposes a structural weakness in how sustainability is often taught. Courses that deliver content through lectures and assessments naturally reward students who master material within the semester’s boundaries. But lifelong learning is a metacognitive skill, a habit of self-directed inquiry that must be deliberately cultivated through open-ended projects, reflective practice, and iterative feedback. The intermittent deterioration suggests that whenever course design drifted toward content delivery and away from skill cultivation, the fragile competencies degraded first. Educational quality, the data implies, is not a fixed attribute of a program but a moving target that decays without active maintenance.

From this diagnosis, the researchers constructed their principal contribution: a comprehensive CQI framework that integrates global certification standards and stakeholder expectations through iterative feedback loops, data-driven analytics, and structured stakeholder involvement. The framework operates as a closed circuit. Learning outcome data flows in from assessments and feedback platforms, gets analyzed against international benchmarks and industry needs, generates targeted curriculum interventions, and then the cycle repeats with the next cohort measuring whether the intervention worked. It is, in essence, a control system for education, applying the same engineering logic of measurement, correction, and verification that civil engineers apply to the structures they build.

What elevates the framework beyond a local fix is its explicit alignment of three traditionally separate domains: curriculum design, sustainability competencies, and industry requirements. Universities frequently design curricula in academic isolation, accreditors evaluate against abstract standards, and employers complain that graduates arrive unprepared, with little mechanism connecting the three conversations. The Malaysian team’s model forces them into a single feedback architecture, ensuring that what is taught, what is measured, and what the profession actually needs remain coherent. The authors position the result as an evidence-based, adaptable model that other institutions can transplant, a claim strengthened by the framework’s reliance on data structures, such as semester-level outcome tracking and standardized feedback instruments, that most universities already collect but rarely integrate.

The study also carries quiet methodological lessons for the growing field of education research. By mining existing institutional data, pseudonymized marks, feedback records, and quality reports, the researchers produced findings of international significance without a single new data collection instrument, an approach the university’s ethics review waived for formal approval precisely because it involved de-identified secondary data handled in compliance with Malaysia’s Personal Data Protection Act. The implication is provocative: most universities are already sitting on a decade of CQI reports and feedback archives that could yield exactly this kind of diagnostic insight, if anyone thought to analyze them as a longitudinal dataset rather than filing them as bureaucratic paperwork.

The broader stakes extend well beyond one Malaysian campus. Sustainability management education is expanding rapidly across engineering schools worldwide, driven by accreditation mandates, employer demand, and the sheer urgency of climate adaptation. Yet the field has lacked rigorous, longitudinal evidence about whether these courses actually produce the competencies they promise. This study supplies that evidence, and its verdict is nuanced: the social and professional dimensions of sustainability education appear to be working, while the cognitive engines of independent inquiry and adaptive problem-solving require deliberate redesign. For the engineers who will rebuild the world’s infrastructure under climate pressure, the difference between those two outcomes may determine whether the sustainability transition is led by professionals who keep learning, or by ones who stopped at graduation.

Subject of Research: Continuous quality improvement framework for evaluating learning outcomes in sustainability management education for civil engineers

Article Title: Establishing a comprehensive continuous quality improvement framework for evaluating learning outcomes in sustainability management education

Article References: Kamal, N. A., Apandi, N. M., Daud, N. R., & Rahman, N. A. (2026). Establishing a comprehensive continuous quality improvement framework for evaluating learning outcomes in sustainability management education. Discover Sustainability. https://doi.org/10.1007/s43621-026-04633-4

Image Credits: AI Generated

DOI: 10.1007/s43621-026-04633-4

Keywords: sustainability education, civil engineering, continuous quality improvement, learning outcomes, lifelong learning, problem-solving, curriculum design, Malaysia, master's education, student feedback, accreditation standards, educational analytics

The core problem the researchers confronted is one that has frustrated structural engineers for decades. When damage appears in a steel structure, it rarely announces itself with dramatic changes in global vibration behavior. Instead, the dynamic response of the structure shifts in ways so subtle that they can be drowned out by measurement noise, environmental variability, and the sheer complexity of the signal itself. Traditional damage identification methods, which often rely on extracting specific features from vibration data such as natural frequencies or mode shapes, struggle when the damage is localized and the signal-to-noise ratio is poor. The research team’s answer was to stop treating vibration data as a stream of numbers and start treating it as an image, allowing the full pattern-recognition power of modern deep learning to be brought to bear on the problem.

The transformation from signal to image is accomplished through a technique called a Markov transition field, or MTF. To understand what this does, imagine taking a long vibration recording and dividing its amplitude range into a finite number of bands, a process the researchers carried out using approximate equal-frequency binning so that each band contains roughly the same amount of data. Every moment in the recording is then assigned to one of these states. The MTF image is built by asking, for every pair of time points in the signal, how likely is it that the system transitions from the state at one moment to the state at the other. The result is a two-dimensional grid, an image in which the hidden temporal dynamics of the vibration are laid bare as spatial patterns. Subtle damage signatures that would be nearly invisible in a raw time series become visible textures that a convolutional neural network can learn to recognize.

But the researchers did not stop at simply converting data into pictures. They recognized that not all parts of an MTF image are equally informative, and that a network capable of deciding where to look would outperform one that treats every pixel with equal weight. Their solution was to embed a mechanism called coordinate attention, or CA, directly into the residual blocks of a deep convolutional network. Coordinate attention is a refinement of the broader family of attention mechanisms that have transformed machine learning in recent years. Rather than compressing spatial information into a single global descriptor, coordinate attention decomposes the feature maps along one-dimensional directions, performing coordinate pooling separately along the height and width axes. This allows the network to jointly calibrate spatial and channel information, essentially teaching it both what features matter and where in the image those features are located.

The backbone of the framework is ResNet-50, a fifty-layer residual network whose skip connections allow gradients to flow through very deep architectures without vanishing. By weaving coordinate attention into the residual blocks, the researchers created a hybrid that combines the representational depth of a proven architecture with the selective focus of attention. The network was validated on two widely respected experimental platforms: the Qatar University Grandstand Simulator, known as QUGS, and the IASC-ASCE benchmark structure, a standardized testbed that has served the structural health monitoring community for years. These are not toy problems. They are physical structures with realistic geometry, realistic boundary conditions, and realistic damage scenarios, which makes the results far more meaningful than demonstrations on synthetic data alone.

The comparative experiments were thorough and revealing. The researchers pitted the Markov transition field against three other image-encoding methods: the Gramian angular difference field, which transforms time series into a matrix of angular relationships; the recurrence plot, which visualizes recurring patterns in dynamical systems; and the short-time Fourier transform, the classical tool for viewing a signal in both time and frequency. Across the benchmarks, MTF generally outperformed all three alternatives, suggesting that the probabilistic transition structure it captures is particularly well suited to the kind of subtle, local changes that damage produces in vibration data. This finding alone offers practical guidance to engineers choosing how to preprocess monitoring data for machine learning pipelines.

The attention mechanism itself also proved its worth in head-to-head comparisons. Under noise-free conditions, the full framework, ResNet-50 enhanced with coordinate attention, achieved an identification accuracy of 94.5 percent. More importantly, the model demonstrated strong robustness under noisy conditions, which is where many competing approaches falter. Real-world sensor data is never clean; it carries electrical interference, environmental fluctuations, and quantization artifacts. A damage detection system that only works in the laboratory is of limited value, so the noise robustness results may prove to be the most consequential finding of the study. The researchers also compared their approach against baseline models, confirming that the combination of MTF encoding and coordinate attention delivers benefits that neither component achieves alone.

One of the most compelling aspects of the study is the feature visualization analysis, which offers a window into what the network actually learns. The results show that coordinate attention suppresses irrelevant redundancy in the MTF images through coordinate-aware spatial modeling. In plain terms, the attention mechanism learns to ignore the parts of the image that carry no diagnostic information and to concentrate on the regions that distinguish damaged from undamaged states. The consequence is visible in the learned feature space: the network produces highly compact and nearly non-overlapping feature clusters, with the best discriminability among the configurations tested. When a model’s internal representations cleanly separate the classes it must distinguish, it is a strong sign that the model has learned something genuinely meaningful about the physics rather than memorizing artifacts of the training data.

The implications of this work extend well beyond the laboratory benchmarks on which it was validated. Civil infrastructure around the world is aging, and the cost of inspecting bridges, buildings, and industrial facilities by traditional means is enormous. Continuous monitoring systems generate torrents of vibration data, but the bottleneck has always been turning that data into actionable decisions about safety and maintenance. A framework that can automatically identify local damage from raw sensor streams, with high accuracy and resilience to noise, moves the field closer to that goal. The open-access publication of the study under a Creative Commons license also means that the methodology is freely available to researchers and practitioners worldwide, potentially accelerating adoption and further refinement.

There remain, of course, questions that future work must address, as with any method validated on two experimental platforms. Real structures face temperature swings, loading variability, and sensor degradation that no benchmark can fully reproduce. Yet the direction is clear and the results are striking. By teaching a deep neural network to look at vibration data as an image and to pay attention to exactly the right places, Xie, Cui, and Lu have shown that the faintest structural whispers can be amplified into clear, actionable signals. As cities grow taller and infrastructure grows older, the buildings that surround us may soon be watched over by algorithms that notice a crack long before any human eye could, and that quiet vigilance may one day be the difference between routine maintenance and catastrophe.

Subject of Research: Deep learning-based structural damage detection using Markov transition field image encoding and coordinate attention on structural health monitoring vibration data

Article Title: An attention-enhanced residual network framework for structural damage detection based on structural health monitoring data

Article References: Xie, B., Cui, J., & Lu, N. (2026). An attention-enhanced residual network framework for structural damage detection based on structural health monitoring data. Scientific Reports. https://doi.org/10.1038/s41598-026-73433-7

Image Credits: AI Generated

DOI: 10.1038/s41598-026-73433-7

Keywords: structural health monitoring, damage detection, deep learning, coordinate attention, Markov transition field, ResNet-50, vibration analysis, convolutional neural network, civil engineering, machine learning, structural dynamics, noise robustness

The core difficulty is class imbalance. In ordinary concrete CT images, voids — the pores and microcracks that govern how water, chlorides and gases migrate through a structure — occupy only about 3.2 percent of the pixels. Cement paste fills roughly 64.4 percent and aggregate the remaining 32.4 percent. When a three-class network is trained on such skewed data, the optimizer can satisfy itself by simply predicting the dominant classes everywhere. The study’s own controlled experiment makes this vivid: a single-stage Vision Transformer–U-Net hybrid, trained on exactly the same data split, preprocessing and loss functions as the proposed method, collapsed onto the majority aggregate class. It achieved just 31.81 percent overall accuracy and a mean Intersection over Union of 10.61 percent, with F1-scores of essentially zero for both void and paste. In other words, the model learned to call nearly everything aggregate and nothing else.

The proposed alternative decomposes the problem along the physical hierarchy of the material itself. Stage 1 is a binary task: separate void from non-void, exploiting the fact that air and cracks attenuate X-rays weakly and appear dark. Stage 2 then classifies the remaining pixels as paste or aggregate, with void pixels masked out of the loss entirely through an ignore index so they cannot contaminate the second stage’s learning. At inference, the two predictions are merged: anything Stage 1 calls void becomes void, and everything else receives the Stage 2 verdict. Each stage therefore faces a milder imbalance than the original three-class problem, and void no longer competes with paste–aggregate boundaries for the network’s attention during optimization.

Both stages share the same backbone: a Vision Transformer encoder (ViT-B/16, pretrained on ImageNet) paired with a convolutional U-Net-style decoder. The choice is deliberate. Self-attention lets the encoder aggregate long-range context across a 224-by-224-pixel patch, useful when distinguishing fine pores from large aggregate grains, while the upsampling decoder reconstructs dense, full-resolution masks. Because the Vision Transformer produces a flat sequence of patch tokens rather than multi-scale feature maps, the decoder omits encoder skip connections and instead applies four bilinear upsampling blocks that rebuild the image from a 14-by-14 feature grid. Each stage carries roughly 86.6 million trainable parameters, all fine-tuned end-to-end on concrete patches with no concrete-specific pretraining.

Training was equally careful about imbalance. Each stage optimizes a weighted blend of cross-entropy, focal, Dice and IoU losses, with coefficients of 0.5, 0.25, 0.15 and 0.10 respectively. Stage 1 up-weights void pixels fifteenfold; Stage 2 mildly favors aggregate. The optimizer is AdamW with a learning rate of 5×10⁻⁵ under cosine annealing, early stopping on validation loss, and mixed-precision training in PyTorch. Crucially, the single-stage baseline received the same recipe, including void-upweighted losses — showing that reweighting alone, a common remedy in the literature, could not restore minority-class performance under this extreme skew. The gains come from task structure, not hyperparameter tuning.

The results on 151 held-out test patches are striking. The two-stage pipeline reached 94.58 percent overall accuracy, 87.50 percent mean IoU and 93.31 percent mean Dice, with class-wise F1-scores of 0.92 for void, 0.96 for paste and 0.92 for aggregate. Void recall hit 96.4 percent — the number that matters most, since missed pores translate directly into underestimated porosity and unreliable durability predictions. Bootstrap resampling over the test patches showed tiny standard deviations, with overall accuracy at 94.90 percent plus or minus 0.07 percent, indicating the performance is stable rather than a lucky draw. An ablation confirmed that removing either stage causes substantial degradation: Stage 2 alone cannot find voids at all, and Stage 1 alone cannot separate aggregate from paste.

The study also stress-tested the framework beyond its algorithmic training labels. On a second corpus of 82 images with human-prepared masks, held out twelve at a time for testing, the two-stage model outperformed a traditional enhanced-mask baseline on both 8-bit inputs and percentile-normalized 16-bit inputs, with the two encodings performing nearly identically on overall metrics. A full-slice demonstration on an unseen CT image predicted phase fractions of 3.64 percent void, 59.38 percent paste and 36.98 percent aggregate — closely tracking the training distribution and suggesting the pipeline can plausibly map whole specimens when preprocessing matches training assumptions. The authors are careful, however, to frame this as transfer plausibility rather than certified generalization.

The error analysis is refreshingly honest about where the method still struggles. The largest residual confusion lies at paste–aggregate interfaces, where similar X-ray attenuation and partial-volume mixing produced more than 365,000 swapped pixels in the test confusion matrix. Thin voids and microcracks are sometimes absorbed into paste, and beam-hardening streaks near dense aggregate can trigger false void calls. Because the merge rule forces every Stage 1 void prediction to remain void, errors cascade: the 3.6 percent of void pixels missed by Stage 1 can never be recovered, and false voids bypass Stage 2’s judgment entirely. For engineering use, the authors note, bulk porosity and aggregate grading from well-resolved regions are far more trustworthy than interface statistics or sub-resolution pore counts without local audit.

The limitations define a clear research agenda. All primary data come from a single specimen, one scan and one mix design, with patch-level rather than slice-level splitting, so neighboring windows from the same CT slice can straddle train and test folds. The reference labels themselves are algorithm-generated pseudo-labels, verified against manual annotation on only a 57-patch subset (88.64 percent pixel agreement), and the manual-label external check rests on just twelve test images. Inference is strictly two-dimensional, with no inter-slice consistency, and the two forward passes roughly double the computational cost of a one-pass model — 27.9 versus 14.6 milliseconds per patch on an RTX 2060, acceptable for offline analysis but not real-time deployment.

Even within those bounds, the study delivers a lesson that resonates far beyond concrete. When one class is vanishingly rare, the fix may not be a cleverer loss or a bigger model but a restructuring of the task itself — teaching the network to find the rare thing first, and only then sort the rest. As infrastructure agencies worldwide push toward digital twins of aging bridges, tunnels and nuclear containment structures, the microscopic pores inside concrete are where durability is decided. A two-stage trick that finally makes those pores visible to machines could quietly become one of the more consequential pieces of AI in civil engineering.

Subject of Research: Two-stage deep learning segmentation of void, paste and aggregate phases in concrete X-ray CT images under severe class imbalance

Article Title: Two-stage deep learning for segmenting in concrete X-ray CT images

Article References: Wang, Y. (2026). Two-stage deep learning for segmenting in concrete X-ray CT images. Results in Engineering, 32, Article 113430. https://doi.org/10.1016/j.rineng.2026.113430

Image Credits: AI Generated

DOI: 10.1016/j.rineng.2026.113430

Keywords: concrete, X-ray computed tomography, deep learning, image segmentation, Vision Transformer, U-Net, class imbalance, porosity, cement paste, aggregate, civil engineering, digital twin

The research, led by R. Daniel Verdugo-Siqueiros with Luis Pellegrin and Jose Angel Gonzalez-Fraga, is explicitly data-centric rather than model-centric. Instead of inventing yet another neural architecture, the authors asked a different question: what if the real obstacle to automated crack inspection is not the detector but the scarcity of well-labeled examples spanning diverse materials and environmental conditions? Their answer was to exploit the zero-shot generalization abilities of foundational vision models, which have been trained on such vast image collections that they can perform tasks they were never explicitly taught, and to augment limited real training data with synthetic examples generated by artificial intelligence itself.

At the heart of the pipeline sits Segment Anything Model 2, or SAM2, Meta’s celebrated segmentation system capable of isolating objects in images and videos from simple prompts. SAM2 provides the framework’s precise, prompt-driven segmentation capability, tracing the often hairline, branching geometry of cracks that would defeat conventional thresholding methods. But SAM2 alone cannot know where to look. That role falls to Florence2, a unified vision foundation model developed at Microsoft Research, which supplies robust feature detection and localization, effectively telling the segmentation module which regions of a photograph deserve scrutiny. The pairing mimics a human inspector’s workflow: first scan the surface for suspicious marks, then examine each candidate closely.

The third component is perhaps the most surprising. Stable Diffusion XL, the high-resolution image synthesis model best known for generating photorealistic artwork, is repurposed here for two jobs: super-resolution enhancement of real inspection images and the generation of synthetic crack examples to flesh out sparse training sets. By upscaling imagery before analysis, the system preserves the fine detail that thin cracks demand. By manufacturing plausible synthetic cracks, it sidesteps the annotation drought that has plagued the field, where datasets covering concrete, asphalt, masonry and ceramics simultaneously are nearly nonexistent.

The numbers reported in the paper are striking for a system built largely from off-the-shelf components. Detection accuracy and precision both reach 92.3 percent, with recall of 89.7 percent and an F1-score of 91.0 percent, meaning the framework misses relatively few genuine cracks while raising few false alarms. The segmentation module, which must trace exact crack boundaries rather than simply flag their presence, starts at a mean Intersection over Union of 63 percent, a Dice coefficient of 77.2 percent, and a boundary F1-score of 68.1 percent. Those boundary metrics matter enormously in practice, because crack width is a key input to structural condition ratings, and sloppy edges translate directly into misjudged severity.

What elevates the work beyond a static benchmark is its use of active learning. Across three successive learning cycles, the segmentation performance climbs from an initial mIoU of 63 percent to 72 percent. In an active learning loop, the model flags the examples it is least confident about, human annotators label only those, and the model retrains on the enriched set. This targeted feedback means each hour of expert labeling yields maximum benefit, progressively refining the model exactly where it struggles, such as shadows, stains, and surface textures that masquerade as cracks. The authors describe this as a reflection of progressive refinement, and the trajectory suggests further gains with additional cycles.

The broader context explains why this matters. The paper’s own references trace a decade of effort, from classical image-processing techniques built on Otsu thresholding and differential images, through deep convolutional networks like DeepCrack and transformer-based systems such as Crackformer, to recent fine-tuning of vision foundation models for civil infrastructure. Each generation improved sensitivity, but each also inherited the same fundamental constraint: models are only as good as the labeled cracks they see. Surveys of crack detection technologies from 2022 and 2023 document the field’s rapid growth alongside persistent generalization failures, where a model trained on one pavement type stumbles on another. A framework that leans on zero-shot capability and synthetic augmentation is a direct strike at that weakness.

Computationally, the system is refreshingly lean. Processing high-resolution images in approximately 2.3 seconds on a standard graphics card means the pipeline could plausibly run on the kinds of hardware mounted on inspection drones, robots, or even field laptops, rather than being confined to cloud clusters. That opens the door to comprehensive monitoring programs in which bridges, tunnels, dams and highway networks are photographed routinely and analyzed automatically, with cracks identified, measured and tracked over time. The authors report robust performance across diverse materials and environmental conditions, the variability that has historically broken narrower systems. Data and models are available on request, and the team maintains a public GitHub repository, smart-infrastructure-inspector, dedicated to the work.

The stakes are not abstract. The collapse of the I-35W bridge in Minneapolis in 2007 and the 2021 closure of the I-40 bridge over the Mississippi River after a cracked tie girder was discovered put the cost of undetected deterioration on national display. Probability-of-detection studies for visual inspection of steel bridges have shown how inconsistent human inspection can be, and maintenance economics research repeatedly finds that early, inexpensive intervention prevents catastrophic, budget-breaking repairs. An automated inspector that never tires, never skips a hairline fracture in poor lighting, and improves with every labeled example could shift infrastructure management from reactive crisis response toward genuine predictive maintenance.

There are, of course, caveats. The reported metrics come from the authors’ own evaluation, and the paper does not yet demonstrate deployment at the scale of an entire highway network or a fleet of aging bridges. Synthetic data, however cleverly generated, risks embedding biases if the diffusion model’s imagined cracks diverge from real failure modes, and human annotators remain essential to the active learning loop. Still, the conceptual move is hard to dismiss: rather than waiting for a perfect crack dataset that may never exist, the framework manufactures its own training material and borrows perception from models trained on billions of images. As foundational vision models continue to mature, the authors’ integration strategy, pairing detection, segmentation and generation in a single data-centric loop, may well become the template for automated inspection across civil engineering, and a quiet but meaningful step toward infrastructure that tells us it is failing before it does.

Subject of Research: Automated crack detection and instance segmentation in construction materials using foundational vision models

Article Title: A data-centric framework for automated crack instance segmentation in construction materials via integration of foundational vision models

Article References: Verdugo-Siqueiros, R. D., Pellegrin, L., & Gonzalez-Fraga, J. A. (2026). A data-centric framework for automated crack instance segmentation in construction materials via integration of foundational vision models. Multimedia Tools and Applications, 85(10), Article 807. https://doi.org/10.1007/s11042-026-21964-6

Image Credits: AI Generated

DOI: 10.1007/s11042-026-21964-6

Keywords: crack detection, instance segmentation, foundation models, SAM2, Florence2, Stable Diffusion XL, infrastructure monitoring, computer vision, active learning, synthetic data, civil engineering, concrete

Developed by Ji’an Liao, Zifa Wang, Dengke Zhao, Jianming Wang, Zhaoyan Li, and Siran Yang, SHM-EM addresses a deceptively simple failure mode: machine learning models used for forecasting demand their inputs in a fixed order with fixed columns, while monitoring platforms store data in vendor-specific tables with inconsistent fields, units, and point identifiers. The result is brittle glue code that is hard to verify, harder to reuse, and nearly impossible to audit when something goes wrong. Existing standards such as the Open Geospatial Consortium’s SensorThings API standardize how sensors and observations are described and queried, and complex-event-processing research supplies tools for windows, conditions, and event generation, but neither specifies how a persisted forecast should be inspected, aligned, and conditionally admitted into an official event workflow. That boundary is precisely where the new framework concentrates its effort.

Architecturally, SHM-EM is organized into four layers: a Vue-based user interface, Spring Boot application services, a Python forecasting runtime named PIT_PRE that runs fixed-version models, and a MySQL persistent store holding observations, model registrations, predictions, events, and responses. The authors deliberately keep the Python modelling ecosystem separate from the Java services, which isolates inference from page-access frequency and prevents model dependencies from leaking into the back end. Formal event creation is reserved exclusively for back-end services, a design choice that reflects the framework’s central concern: the difference between looking at a forecast and acting on one must be enforced by the software itself, not by user discipline.

The first of the framework’s three core mechanisms is a versioned, engineering-semantic data-model contract. This contract binds monitoring objects, source fields, ordered input features, fixed-version models, and prediction targets into a single inspectable registration. Cryptographic bundle hashes cover model weights, preprocessors, scripts, parameters, and runtime metadata, while feature mappings retain input order, source identity, units, conversion versions, and target roles. When anything drifts, whether the bundle, the schema, the contract, or the feature ordering, PIT_PRE rejects the run outright. Before inference, observations are aligned to a registered three-minute grid using backward as-of matching within one cadence, followed by linear interpolation and boundary fill, with signed time offsets and fill diagnostics recorded. Unresolved required inputs are rejected rather than silently imputed, and data freshness is checked separately.

The second mechanism is the project future state, a software-level representation of synchronized forecasts and risk summaries. In the reference deployment, six Transformer-CNN models forecast deep horizontal displacements in two directions, earth-pressure strain and pressure, groundwater elevation, and surface settlement. Five of the models consume 114 input features each while the settlement model uses 164, and together they produce 124 target channels across 40 future steps at three-minute intervals. All models share a single prediction origin and future timeline, so a project manager sees one coherent 120-minute view of predicted risk rather than six disconnected model outputs. A policy-bound aggregation algorithm then evaluates rule-level conditions step by step, activates severity levels when consecutive-step requirements are satisfied, assigns the highest activated severity, and merges forecast risk with any open risk from already-observed events.

The third mechanism, and arguably the most consequential, is the controlled transition from forecasts to formal engineering events. After a prediction batch is stored, SHM-EM validates it and prepares a unified engineering series in which observed and forecast values share object, metric, time, engineering-value, unit, and provenance fields. A gate then checks batch status, required runs and steps, timing, quality, bundle hashes, and the recomputed integrity of persisted results. Only then are two independent paths exposed. Evaluate returns candidate events and calculation snapshots, storing an audit run without creating any formal event or response record. Execute, by contrast, never consumes a stored evaluation result; it reloads the series, persists a fresh execution-gate result, revalidates rule and unit semantics, and creates the formal event, response workflow, and provenance chain only when every condition is satisfied. This separation means that a mutation of data after evaluation cannot slip a stale decision into formal execution, a failure path the authors explicitly test.

Provenance receives equally careful treatment. Every formal event can be traced to its rule version, prediction batch, model and input hashes, the exact input window, the exceedance details including lead time, peak value, and consecutive steps, and a result hash. In the published reproduction, a captured event resolves to rule version 2, a 40-step batch, a settlement model run, and an input window spanning late December 2024 into the first minutes of 2025, with the machine-readable trace retaining a peak settlement of roughly 9.43 millimeters. Such traces transform what is usually an opaque alerting pipeline into an auditable record that a reviewer can reconstruct end to end.

The evidence base for these claims is unusually concrete. A public reference case built on 2,464 de-identified, time-shifted observations from nine excavation points verifies the software chain without claiming predictive accuracy, since confidentiality prevents release of the complete training data. A 15-case validation matrix combined one positive control, twelve injected failure paths, and two input controls; every expected blocked case produced zero formal business records, while audit writes remained permitted. Scripted reproduction families passed in full, including 55 back-end tests, 13 PIT_PRE tests, and the end-to-end reference workflow, and a Docker Compose route reproduced the entire six-model, 4,960-record pipeline. A synthetic bridge fixture then demonstrated reuse: adding three stations, twelve instruments, and new mappings required no changes to back-end, front-end, forecasting, or event code, with missing mappings correctly rejected. Runtime measurements on a warm-cache Windows machine showed a full six-model prediction batch completing in a median of about 16.8 seconds, with the execution gate itself taking only a few hundred milliseconds.

The authors are candid about limits, which strengthens rather than weakens the contribution. The excavation case and synthetic fixture do not validate predictions for other structure types, the point forecasts provide no calibrated uncertainty or safety confidence, and MySQL is the only tested back end behind a 50,000-row application-level gate cap. Cross-platform reproduction revealed that while input and model-contract hashes matched between Windows and Ubuntu, normalized prediction hashes differed slightly, with a maximum persisted absolute difference of about 0.00285, so Windows remains the exact-output reference. The public baseline lacks authentication, and the security documentation requires TLS, external identity and role-based access control, separate Execute privileges, least-privilege database accounts, and protected audit logs. Most importantly, the authors state plainly that forecasts must never be the sole basis for automated safety decisions.

Even within those boundaries, the significance of SHM-EM lies in what it makes routine. Researchers can register compatible model versions, thresholds, and rule settings without rewriting acquisition, forecasting, or event-handling workflows. After one-time registration, non-specialist users work through a web interface to inspect observed and forecast series, review threshold exceedances and lead times, and, crucially, distinguish a candidate event from a formally executed one, with eligibility or blocking reasons, model versions, input windows, and hashes all visible. Released under the MIT license with locked dependencies, immutable model cards, and a SHA-256-verified archive, the framework offers the structural health monitoring community something rarer than another forecasting algorithm: a disciplined, reproducible pathway by which a prediction earns the right to become an official record of engineering concern.

Subject of Research: A forecast-aware event management software framework for structural health monitoring of civil engineering projects

Article Title: SHM-EM: a forecast-aware event management framework for heterogeneous engineering monitoring

Article References: Liao, J., Wang, Z., Zhao, D., Wang, J., Li, Z., & Yang, S. (2026). SHM-EM: a forecast-aware event management framework for heterogeneous engineering monitoring. SoftwareX, 36, Article 103082. https://doi.org/10.1016/j.softx.2026.103082

Image Credits: AI Generated

DOI: 10.1016/j.softx.2026.103082

Keywords: structural health monitoring, SHM-EM, machine learning forecasting, Transformer-CNN models, sensor data integration, event management, provenance, reproducibility, open-source software, excavation monitoring, data-model contract, civil engineering

The core problem the researchers set out to solve is one that plagues nearly every large infrastructure domain: heterogeneity. Levee records are maintained by different organizations using different vocabularies, formats, and levels of detail. A levee system may be described one way in a federal database, another way in a state GIS layer, and yet another way in an engineering inspection report. This fragmentation limits interoperability, undermines semantic consistency, and ultimately hampers risk-informed decision-making. When flood managers cannot reliably connect a culvert record to the levee segment it penetrates, or trace a monitoring asset back to the system it serves, the consequences can be measured in degraded situational awareness during emergencies.

The team’s solution centers on transforming normalized levee datasets into what they call a Levee System Knowledge Graph, or LSKG. A knowledge graph represents information as entities, such as levee systems, segments, structural components, and monitoring assets, connected by meaningful relationships. Rather than treating records as isolated rows in a table, the graph encodes the actual structure of the world: this segment is part of that system, this component is located on that segment, this sensor monitors that structure. The researchers aligned their normalized datasets with the National Levee Database Ontology, abbreviated NLD-O, a formal schema that defines the classes and relationships relevant to levee infrastructure in the United States. Ontology alignment is what gives the graph its semantic spine, ensuring that the same concept is represented the same way regardless of which source dataset it came from.

The choice of ontology was not arbitrary. The National Levee Database, maintained by the U.S. Army Corps of Engineers, is the country’s authoritative repository of levee information, and the datasets analyzed in the study are publicly available from it. By anchoring the knowledge graph to an ontology derived from this national resource, the framework inherits a vocabulary that flood-risk professionals already use, which matters enormously for adoption. The authors drew on established ontology engineering practices, including the Common Core Ontologies suite and the Basic Formal Ontology, both of which provide rigorously specified mid-level frameworks for representing entities, processes, and their relationships. Tools such as Protégé, the long-standing ontology editor developed at Stanford, support this kind of formal knowledge modeling.

Building the graph, however, is only half the challenge. The real innovation lies in how the framework answers questions. The researchers constructed a graph-addressable semantic corpus and deployed a hybrid retrieval strategy that combines three complementary techniques. The first is TF-IDF vectorization, a classical information retrieval method that dates back to Karen Spärck Jones’s foundational 1972 work on term specificity. TF-IDF weighs terms by how frequently they appear in a document relative to how common they are across the whole corpus, allowing the system to find text passages that are statistically relevant to a query. The second technique is graph reasoning, which exploits the explicit structure of the knowledge graph itself, traversing relationships between entities to answer questions that no single document could answer alone. The third is large language model assistance, in which an LLM interprets the retrieved evidence and composes a natural-language response.

This hybrid architecture is significant because each component compensates for the weaknesses of the others. Pure vector retrieval can find text that looks relevant but miss answers that require multi-hop reasoning across the graph. Pure graph queries are precise but demand structured input that casual users cannot provide. Large language models are fluent but notoriously prone to hallucination when asked to reason about facts they have never seen. By grounding the LLM in retrieved evidence from both the semantic corpus and the graph, the framework follows the retrieval-augmented generation paradigm introduced by researchers at Facebook AI in 2020, while extending it with graph-native reasoning. Crucially, the design preserves traceability: every answer can be linked back to the specific infrastructure records that support it, a property that is essential in engineering contexts where accountability matters.

To evaluate the system rigorously, the authors went beyond simply demonstrating that a knowledge graph could be built. They introduced a quantitative benchmark of ontology-aware levee infrastructure questions spanning four task categories: entity retrieval, which tests whether the system can identify specific objects; attribute retrieval, which tests whether it can report properties of those objects; relationship reasoning, which tests whether it can navigate connections between entities; and aggregation tasks, which require counting, summing, or otherwise computing over sets of records. Performance was measured using Exact Match, which requires the response to match a reference answer precisely, and Contains Match, which accepts answers that contain the correct value, along with response latency.

The results were remarkable in their consistency. The framework achieved 95.16 percent on both Exact Match and Contains Match, indicating that when the system answered correctly, it did so with precision rather than vague approximation. For a domain where a misidentified levee segment or an incorrect culvert count could have safety implications, that level of exactness is the difference between a research demo and a decision-support tool. The benchmark itself is a contribution, since the authors note that evaluating GraphRAG systems on domain-specific, ontology-aware questions remains rare, and they explicitly call for broader validation using additional datasets and independent benchmarks.

The implications extend well beyond levees. Flood risk is escalating as climate change drives non-stationary hydrology, a shift famously summarized in a 2008 Science paper declaring that stationarity is dead in water management. Agencies including FEMA and the U.S. Army Corps of Engineers have emphasized that levee safety depends on integrating inspection records, geospatial data, and monitoring information, and recent research has explored UAV-based thermal imaging for seepage detection, InSAR satellite deformation monitoring, and machine learning for culvert inspection. All of these technologies generate yet more heterogeneous data, which makes a semantic integration layer arguably a prerequisite for any intelligent flood-risk governance. A knowledge graph that can absorb sensor observations, geospatial reasoning, and temporal information, as the authors envision in future extensions, could become the connective tissue for next-generation infrastructure intelligence.

The study also offers a template for how scientific domains broadly might tame their own data chaos. Geoscience knowledge graph efforts, from earthquake emergency support frameworks to large collaborative programs, have been converging on the same insight: large language models become dramatically more reliable when anchored to formally structured, domain-specific knowledge. The levee work demonstrates the full pipeline, from ontology development and data normalization through graph construction in a platform like Neo4j, to hybrid retrieval and benchmark evaluation. Supported by the National Geospatial-Intelligence Agency through the Oak Ridge Institute for Science and Education, and grounded in publicly available National Levee Database records, the framework shows that trustworthy, traceable, natural-language access to critical infrastructure knowledge is no longer a distant aspiration but an engineering achievement with a measured score of 95.16 percent to prove it.

Subject of Research: An ontology-driven GraphRAG framework integrating levee infrastructure data into a knowledge graph for semantic querying and flood-risk decision support

Article Title: Hybrid GraphRAG framework for semantic integration and querying of levee infrastructure data

Article References: Davarpanah, A., Scully, E. J., Lipscomb, M. R., & Morris, A. L. (2026). Hybrid GraphRAG framework for semantic integration and querying of levee infrastructure data. Earth Science Informatics, 19(11), Article 185. https://doi.org/10.1007/s12145-026-02234-3

Image Credits: AI Generated

DOI: 10.1007/s12145-026-02234-3

Keywords: levee infrastructure, knowledge graph, GraphRAG, ontology, large language models, flood risk, National Levee Database, semantic integration, information retrieval, TF-IDF, Neo4j, civil engineering

Khanna’s path to the chairmanship was, in a sense, an extended audition. During his fourteen months as interim chair, he undertook what colleagues describe as an unusually systematic listening exercise, meeting extensively with faculty, staff, and students while also engaging the department’s external partners in industry and government. The goal was not ceremonial. He wanted a granular picture of the department’s strengths, its unmet needs, and the opportunities where civil and environmental engineering could grow. Those conversations fed directly into concrete administrative changes, including strengthened departmental operations, more rigorous academic and faculty planning, and the creation of a CEE Honors and Awards Committee designed to formally recognize excellence within the department.

That record of translating consultation into action was central to the case for his permanent appointment. Michele V. Manuel, the U. S. Steel Dean of Engineering, praised Khanna’s interim tenure as a model of responsive leadership. In her assessment, he listened carefully to every constituency the department serves and then converted what he heard into institutional change. She noted that he understands the department deeply, has already strengthened its operations and planning processes, and articulates a clear vision for where civil and environmental engineering at Pitt can expand in education, research, and external engagement. For a school weighing candidates from a national search, that combination of institutional knowledge and demonstrated execution proved decisive.

The search itself was a formal affair, chaired by Sanjeev Shroff and conducted by a committee that drew representation from the department’s faculty, staff, and students, as well as from the Swanson School leadership and the CEE Visiting Committee. That breadth of participation reflected the stakes: the chair of a civil and environmental engineering department sits at the intersection of academic priorities and the practical demands of infrastructure agencies, engineering firms, and municipalities that depend on the department’s graduates and research. The committee’s recommendation of Khanna, following the national search, signaled confidence that an internal candidate offered the strongest vision for the department’s future.

Khanna’s own academic trajectory is unusually broad for a civil and environmental engineer, and it helps explain the interdisciplinary ambitions he brings to the role. He joined the University of Pittsburgh in 2010 and rose through the faculty ranks, being promoted to full professor in 2023. Between 2017 and 2022, he served as the department’s associate chair for graduate studies, a position that placed him in charge of graduate recruitment, admissions, orientation, academic requirements, and program development. He has also chaired faculty searches and sat on committees at both the school and university levels, giving him a comprehensive view of how the institution’s academic machinery operates.

His educational background spans chemical engineering, statistics, and applied mathematics. Khanna earned his bachelor of engineering in chemical engineering from Punjab University before moving to The Ohio State University, where he accumulated three graduate degrees: a master’s in applied statistics and both a master’s and a doctorate in chemical engineering. That quantitative foundation is evident in his scholarship, which treats environmental systems not as isolated problems but as interconnected networks that can be modeled, optimized, and stress-tested. His methodological toolkit includes systems analysis, life-cycle assessment, optimization techniques, and data-driven modeling, all deployed to examine the environmental consequences of emerging technologies and complex engineered systems.

The subject matter of that research reads like a map of the challenges confronting infrastructure in the twenty-first century. Khanna has focused on water and resource recovery, exploring how wastewater and industrial byproducts can be transformed from liabilities into assets. He has examined the circular economy, asking how materials can circulate through the economy repeatedly rather than flowing from extraction to disposal. His work on resilient infrastructure addresses how civil systems can be designed to withstand shocks, whether from climate-driven extremes or other disruptions, and he has investigated emerging energy and water technologies that promise to decarbonize and democratize access to essential services. The through-line is a conviction that environmental performance must be quantified rigorously before it can be improved.

The scholarly record supporting that conviction is substantial. Khanna has published more than 75 peer-reviewed articles and has served as principal investigator or co-principal investigator on nearly $10 million in externally funded collaborative research, a figure that reflects both the competitiveness of his proposals and the collaborative character of his science. Since 2017, he has served as an editor of Resources, Conservation and Recycling, one of the leading journals in the field of resource sustainability, where he has helped shape the scientific conversation around material flows and circularity. His professional service extends to leadership roles within the Sustainable Engineering Forum of the American Institute of Chemical Engineers, and he participated in the National Academy of Engineering’s Frontiers of Engineering Education Symposium, a program that convenes rising leaders to rethink how engineers are trained.

Now, as permanent chair, Khanna has laid out a forward-looking agenda built on the premise that civil and environmental engineering is inherently interdisciplinary and should be organized as such. Among his stated priorities is a deeper integration of artificial intelligence into the department’s curricula, ensuring that graduates are fluent in the machine learning and data analysis tools increasingly used to design, monitor, and manage infrastructure. He also plans to develop new graduate pathways that connect civil and environmental engineering with public health and computing, formalizing the links between water quality, environmental conditions, and human health outcomes. In parallel, he intends to expand IRISE, the department’s initiative in resilient and intelligent infrastructure, and grow its partnerships, while strengthening research at the nexus of water, the environment, and public health.

The vision, as Khanna himself frames it, is grounded in the everyday ubiquity of the discipline he now leads. Civil and environmental engineering, he notes, shapes the systems that people rely on every single day, from the water that arrives at the tap to the roads and bridges that carry daily commutes. His ambition is to strengthen those systems through interdisciplinary education and research, working alongside the department’s faculty, staff, students, and external partners. In an era when infrastructure must simultaneously adapt to climate change, incorporate intelligent technologies, and serve public health, the University of Pittsburgh has placed its confidence in a leader whose entire career has been devoted to quantifying and improving the environmental performance of the systems that hold modern life together.

Subject of Research: Appointment of sustainability researcher Vikas Khanna as Chair of Civil and Environmental Engineering at the University of Pittsburgh Swanson School of Engineering

Article Title: Vikas Khanna named Chair of Civil and Environmental Engineering at Pitt

Article References: Vikas Khanna named Chair of Civil and Environmental Engineering at Pitt. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: Vikas Khanna, University of Pittsburgh, Swanson School of Engineering, civil engineering, environmental engineering, sustainability, life-cycle assessment, circular economy, resilient infrastructure, artificial intelligence, water resource recovery, public health

The core engineering challenge addressed by the researchers is deceptively simple to state and extremely difficult to solve: how do you build a free-swimming robot that can travel far into a confined, water-filled tunnel, navigate reliably without satellite positioning, and return with usable data about the tunnel’s structural condition? Most commercial remotely operated vehicles were designed for open water, harbours or offshore platforms. In a tunnel just a few metres in diameter, propeller wash stirs up clouds of sediment, acoustic navigation signals bounce chaotically off curved walls, and any loss of control could leave a multi-million-dollar robot wedged irretrievably far from the nearest access shaft.

Water conveyance tunnels also present an environment that is hostile in less obvious ways. Flowing water inside an operating tunnel generates currents that can sweep a small vehicle off course, and the tunnel lining itself may include bends, junctions, steep gradient changes and transitions between free-surface and pressurised flow sections. Visibility is frequently near zero because of suspended silt, biological growth and air bubbles entrained by turbulence. The vehicle described in the new work was therefore conceived from the outset as a long-range, confined-space platform rather than an adaptation of offshore robotics, with its hull, propulsion, sensing and navigation systems all shaped by the geometry and hydraulics of conveyance tunnels.

Central to the design is a propulsion and control scheme tuned for low-disturbance operation in narrow passages. Conventional thrusters produce downwash that resuspends sediment and destroys optical visibility, so the vehicle’s thrust configuration and control algorithms prioritise smooth, low-speed transit with minimal wake. The body is typically streamlined to a compact diameter so it can pass through access ports, bends and variable cross-sections that would block larger machines. Redundant thrusters or vectored thrust arrangements allow the vehicle to hold station against tunnel flow while still imaging the walls, a capability that matters because inspection is not simply a matter of driving through a tunnel but of systematically covering every metre of its surface.

Navigation without GPS is the second pillar of the system. Inside a tunnel, satellite positioning is unavailable the moment the vehicle submerges past an access portal, so the robot must estimate its own position using onboard sensors alone. The reported approach relies on combinations of inertial measurement, velocity estimation from onboard instruments such as Doppler velocity logs or flow sensors, and feature matching against the tunnel walls themselves. In a smooth cylindrical conduit, acoustic ranging to the perimeter provides a natural odometry signal: by continuously measuring distances to the wall in multiple directions, the vehicle can infer how far it has travelled and where it sits within the tunnel cross-section, correcting drift that would otherwise accumulate over kilometres of travel.

Perception in murky water is equally demanding, and the study reflects the broader shift in underwater inspection toward sensor fusion. Optical cameras deliver the highest-resolution images of cracks, spalling, joint displacement and rebar exposure in the tunnel lining, but only within a short range and only when the water is clear. Sonar penetrates turbidity but at coarser resolution. By combining optical imagery, acoustic profiling and distance measurements, the vehicle can build a structural map of the tunnel that remains informative even when individual sensors degrade. Onboard lighting must also be engineered carefully, because uneven illumination in a reflective wet surface can wash out the very defects inspectors need to see.

The practical payoff of such a vehicle is substantial. Long-distance tunnels are typically inspected at great cost by dewatering sections of the conduit, a process that interrupts water supply for days or weeks, requires energy-intensive pumping, and can itself stress the structure by changing pore pressures and loads. Robotic in-service inspection would allow operators to assess lining integrity, sediment accumulation, leakage paths and joint condition without taking the tunnel offline. Early detection matters: small lining cracks and localised seepage can progress into major failures that are enormously expensive to repair, particularly in tunnels buried deep underground or crossing difficult terrain where access for heavy machinery is limited.

The work also speaks to a wider trend in civil infrastructure management: the move from periodic, disruption-heavy manual inspection toward continuous, data-rich robotic monitoring. Engineers increasingly describe infrastructure in terms of digital condition models that are updated over time, allowing maintenance to be scheduled on evidence rather than on fixed calendars. A vehicle that can repeatedly traverse a conveyance tunnel, collect consistent imagery and sensor data, and localise its findings precisely along the tunnel axis becomes a mobile sensing node for such models. Over repeated missions, change detection algorithms could highlight newly developed cracks or growing sediment deposits long before they become visible to a human inspection crew working in drained conditions.

Significant engineering hurdles remain before such systems become routine. Energy storage still constrains endurance, since pushing a vehicle against tunnel flow for many kilometres demands sustained power, and battery capacity has not kept pace with the ambitions of long-range missions. Communication is another constraint: radio signals do not penetrate water and rock, so the vehicle must operate with high autonomy, making its own decisions about obstacle avoidance, data quality and abort conditions, with only intermittent contact through tether or acoustic links. Robustness against entanglement, snagging on protrusions, and unexpected obstructions such as gates or debris is essential, because a stranded vehicle in an operating water tunnel could become an obstruction far more costly than the inspection it was meant to perform.

The researchers frame the vehicle as a step toward making confined-water infrastructure inspectable on demand rather than on rare, disruptive occasions. As urban water systems age and climate variability places new demands on conveyance networks, the ability to look inside the pipes and tunnels that sustain cities, without emptying them, may shift from an engineering curiosity to a standard expectation. If the endurance, autonomy and sensing capabilities demonstrated in this work can be scaled and hardened for field deployment, the dark, inaccessible interiors of the world’s water conveyance tunnels could become among the most closely watched infrastructure spaces anywhere, monitored kilometre by kilometre by machines that were built precisely for the environment where humans cannot safely go.

Subject of Research: Development of an unmanned underwater vehicle for autonomous inspection of long-distance water conveyance tunnels

Article Title: An unmanned underwater vehicle for long-distance water conveyance tunnel inspection

Article References: Shen, X., Chu, Z., & Zhu, D. (2026). An unmanned underwater vehicle for long-distance water conveyance tunnel inspection. Communications Engineering. https://doi.org/10.1038/s44172-026-00758-6

Image Credits: AI Generated

DOI: 10.1038/s44172-026-00758-6

Keywords: unmanned underwater vehicle, water conveyance tunnel, infrastructure inspection, autonomous navigation, tunnel lining, robotics, civil engineering, sensor fusion, underwater perception, infrastructure monitoring, unmanned, underwater