Abstract Underground pipelines play a vital role in urban operation and development, and the rapid and accurate detection of their distribution remains an urgent challenge. Ground-penetrating radar (GPR) technology has been widely adopted for underground pipeline detection. Therefore, there is currently a need for an automated solution to detect pipeline categories and measure burial depth and diameter using GPR. This study proposes an automatic recognition method of category, burial depth, and diameter for underground pipeline using GPR based on you only look once (YOLO) v12 and a machine learning algorithm. The method operates through three sequential phases. In the first phase, a well-trained YOLOv12 model detects pipeline signal regions and categories in GPR B-scan images, where the model achieves average precision (AP) scores of 0.649 for plastic–water category, 0.654 for electric-cable category, 0.755 for plastic–gas category, and 0.737 for steel–water category, with a mean average precision of 0.699 at intersection over union thresholds from 0.5 to 0.95 with a step size of 0.05. Then, the second phase applies a hyperbolic feature extraction algorithm to derive hyperbolic equations from the detected pipeline signal regions, and the standard deviations of relative errors are 2.9% for vertex vertical coordinate and 12.4% for vertex curvature. The last phase utilizes a well-trained CatBoost regression model with the independent strategy to measure pipeline burial depth and diameter, where the CatBoost model yields mean absolute error of 0.0154 and 0.0185 m, root mean squared error of 0.0261 and 0.0250 m, mean absolute percentage error of 1.38% and 8.41%, and R -squared scores of 0.9973 and 0.9425, respectively. This study aims to provide valuable references for subsequent research on automatic interpretation methods of underground pipeline information based on GPR.
Abstract We developed a comprehensive hydraulic excavator training simulator to evaluate and train operator skills using both traditional joysticks (JS) and a novel world-space interface (WSI). The simulator features multiple virtual environments modeled after real construction sites, each with varying levels of difficulty to train and assess naive operators. The simulator accurately reproduces the excavator’s motion dynamics, including its hydraulic behavior, mass, and inertia. To simulate the environment dynamics in real time, we employed the Unity3D physics engine to interact with solids (e.g., when removing fallen rocks) and assets with modified shaders to simulate digging or dumping dirt. The novel WSI is a scaled version of the excavator’s linkage, affording a more intuitive control scheme to a naive operator. We conducted two complementary studies: (1) evaluating training effectiveness with traditional joysticks, and (2) comparing performance between joysticks and the WSI. Results showed that naive subjects could improve their performance with joysticks after 7 training days. Naive subjects’ posttraining performance was not significantly different from that of expert operators for simple tasks such as digging and dumping. The use of the WSI resulted in 37% faster cycle times than the joystick interface at first use, closely approximating the posttraining performance of naive operators.
Abstract The paper introduces PyPBEE, an open-source Python library for performance-based earthquake engineering (PBEE). This computational tool is designed to enable researchers and engineers to conduct complex PBEE analyses with enhanced efficiency, flexibility, and accessibility. The library encapsulates a broad range of state-of-the-art computational models and methods required for each phase of PBEE, from seismic hazard analysis to damage prediction and decision-making. PyPBEE aims to streamline the PBEE workflow by offering a user-friendly, programmable interface that currently accommodates probabilistic seismic hazard analysis, ground motion selection, nonlinear time-history analysis, probabilistic seismic demand hazard analysis, and probabilistic seismic damage hazard analysis. One of its key features is the integration of finite element models with randomized parameters, which effectively introduces a crucial probabilistic dimension to PBEE analysis that is often overlooked. It integrates data management functionalities, enabling seamless manipulation and processing of massive, high-dimensional data sets common in earthquake engineering. The paper presents an extensive exploration of the library’s architecture, functionalities, and capabilities. Real-world case studies demonstrate how PyPBEE can facilitate comprehensive, accurate earthquake risk assessments, thus aiding in the design of more resilient infrastructure. The modular design of the library facilitates and encourages community contributions, enabling its ongoing enhancement to reflect emerging trends and advancements in earthquake engineering. By providing a consolidated, intuitive, and extensible platform, PyPBEE empowers researchers, engineers, and decision-makers to perform sophisticated PBEE analyses, enhancing our ability to understand, predict, and mitigate the impact of seismic hazards.
Abstract The construction industry accounts for a significant portion of workplace fatalities in the United States. Health and safety hazards frequently go unrecognized, potentially leading to injuries and death. However, traditional computer vision methods are often limited to specific object detection tasks [e.g., personal protective equipment (PPE) compliance] and their adoption introduces challenges in interpreting complex, contextual hazard scenarios. Furthermore, the scarcity of annotated accident data hinders the development of fully supervised models. Consequently, this manuscript formally characterizes the performance of pretrained multimodal Artificial Intelligence (AI) for automating hazard recognition. Furthermore, the performance of such models is also tested on game engine-based synthetic data, to investigate whether it can be leveraged for data augmentation in order to overcome the scarcity of real-world training examples. To address these questions, this study formally evaluates the effectiveness of AI models over multiple iterations of the models’ architecture for the domain-specific application of automated construction hazard assessment from multimodal inputs. The research presented herein leverages publicly available pretrained vision-language models and evaluates their performance across three core tasks: binary hazard alerts, detailed hazard understanding, and automated citation generation. A dataset of annotated real-world construction images is presented to benchmark the performance of models. The dataset is also made public to provide a standardized means for comparing performance across different models. Results reported herein indicate that leading AI models can achieve high, although not perfect, recognition accuracy without fine-tuning. However, model performance varies across hazard categories, and newer generations show diminishing returns. Consequently, the study introduces and validates high-fidelity, game engine-based synthetic images as a solution. By generating high-fidelity digital images of dangerous activities, synthetic data can help bridge the scarcity of real-world hazard examples. This approach increases opportunities to improve these models with task-specific data, supporting continued progress beyond current limitations.
Abstract This paper proposes a multiagent large language model (LLM)–based system for early-stage building layout planning, which enables flexible design requirement inputs and robust spatial reasoning. Existing generative methods often depend on structured inputs, and perform poorly when faced with incomplete requirements or regulatory constraints. Moreover, although LLMs excel at understanding language, they typically lack the spatial reasoning capabilities required for layout generation. To address these limitations, the proposed system includes five core modules: a requirement interpreter that standardizes design requirements and employs multimodal retrieval-augmented generation (RAG) to fetch similar layouts; a planner module composed of specialized LLM agents such as an architect, accessibility expert, and user experience advocate; a layout executor that transforms planning strategies into bubble diagrams; and an evaluation module that combines rule-based checks and LLM-based feedback. All modules are coordinated by a director module using multichain-of-thought reasoning to iteratively refine the design process. The system supports natural language prompts, layout boundaries, and optional reference layouts, making it suitable for both incomplete and detailed design scenarios. After a bubble diagram is finalized, a diffusion-based generative model produces a complete architectural layout. Experimental results using the Tell2Design data set demonstrated significant improvements in both geometric quality and semantic alignment over a baseline LLM-only system. Case studies of office and multiapartment buildings further validated the system’s adaptability and effectiveness in complex design contexts.
Abstract Construction robots are increasingly capable of performing complex, labor-intensive tasks such as bricklaying, drilling, and autonomous material handling. Using real-time perception and environmental mapping, intelligent systems can operate effectively in unstructured and dynamic site conditions that traditionally demand human expertise. Recent progress in large language models and vision foundation models offers substantial opportunities to strengthen and extend the capability of creating high-level navigational maps for construction robots. Leveraging these advances, we introduce the zero-shot incremental vision–language maps framework, which is a three-dimensional (3D) modeling system that aims to generate semantically rich map representations of construction sites in a zero-shot manner. Our framework combines simultaneous localization and mapping (SLAM), visual-language feature extraction, incremental semantic and instance label fusion, and spatial graph construction to enable a construction robot navigation framework that supports open-vocabulary language queries. Evaluation is performed on the public ConSLAM dataset, and results show that the proposed framework is capable of building a rich 3D map of columns, signs, framework, and barriers in a construction environment.
Abstract Accurate real-time estimation of human activity intensity is essential for diverse applications such as health monitoring, ergonomics, sports science, and adaptive building management. However, existing methods often depend on intrusive wearable sensors, discrete activity classifications, or extensive training datasets, which compromise their practicality and generalizability. To address these gaps, we propose a novel activity intensity score (AIS) framework that provides a nonintrusive and continuous measure of activity intensity by analyzing video data. The proposed method applies pose estimation to video data to extract body landmarks, which are then used to compute kinematic parameters including the angular velocity, angular acceleration, range of motion, peak speed, movement frequency, and rotational energy across defined kinematic chains (e.g., arms, legs, torso). These parameters are then normalized and combined through an optimized weighted summation to produce a continuous activity intensity metric. Experimental validation was conducted with 20 participants performing various activities from low, moderate, and high intensity. Results demonstrated strong correlations between AIS scores and both activity intensity levels (Spearman’s ρ = 0.943 , p < 0.001 ) and participants perceived exertion ratings (Pearson’s r = 0.923 , p < 0.001 ). Statistical comparisons demonstrated that the AIS values effectively discriminate among these three intensity categories (Spearman’s ρ = 0.943 ) and significant group differences confirmed by ANOVA ( p < 0.001 ). Moreover, the AIS exhibited a strong correlation (Pearson’s r = 0.923 ) with self-reported exertion (Borg rating of perceived exertion), indicating consistency with participants’ subjective perceptions. This occupant-invariant and domain-independent method enables robust, real-time measurement of movement intensity for applications ranging from healthcare and workplace ergonomics to sports analytics and adaptive HVAC control.
Abstract Vehicle detection technology is one of the basic and key technologies for realizing intelligent transportation and autonomous driving. However, in real scenes there are effects such as lighting shadows, motion blur, and target occlusion. This paper proposes a You Only Look Once Version 8 (YOLOv8) vehicle detector Re-Diffusion Task-You Only Look Once (RDT-YOLO) based on feature focused diffusion, aiming to meet the challenge of vehicle detection in complex scenes. A new RepGhost cross stage partial effective long-range aggregation network (RGC-ELAN), focusing diffusion dimension-aware (FDDA) pyramid network, and task align dynamic (TAD) detection head were designed based on the original structure. Experimental results show that RDT-YOLO’s F1 Score increased by 6.0% and the mean Average Precision (mAP) increased by 3.9%. Moreover, the calculation parameters of RDT-YOLO were reduced by 23.6%, the model size was reduced by 20.9%, and the running speed reached 66.2 frames per second (FPS). Additional generalization experiments and robustness tests showed that RDT-YOLO has broad application prospects in different scenarios and can provide reliable support for intelligent transportation systems and autonomous driving technologies.
