Despite significant advancements in point cloud analysis, reducing energy consumption and improving robustness remain understudied, largely due to the inherent limitations of Convolutional Neural Networks (CNNs). To address this, we take the cue from the primary visual cortex and propose a Dendritic-Connected Continuous-Coupled Neural Network (DC-CCNN), a novel Brain-Inspired Neural Network (BINN) architecture tailored for point cloud analysis. By leveraging the unique characteristics of point clouds, our design combines discrete and continuous encoding, replacing traditional Multilayer Perceptrons (MLPs) with more efficient and robust BINNs. Our approach substantially improves the performance of Brain-Inspired Neural Networks on point analysis tasks and maintaining performance comparable to state-of-the-art methods. Furthermore, DC-CCNN exhibits enhanced robustness against various point cloud deformations and corruptions. Our experimental results demonstrate that DC-CCNN achieves competitive performance on benchmark datasets, making it a promising alternative to traditional deep learning methods for point cloud analysis. With its high efficiency and robustness, DC-CCNN has the potential for widespread adoption in 3D computer vision, robotics, and autonomous systems.

Sustainability is becoming increasingly critical in the maritime transport, encompassing both environmental and social impacts, such as Greenhouse Gas (GHG) emissions and navigational safety. Traditional vessel navigation heavily relies on human experience, often lacking autonomy and emission awareness, and is prone to human errors that may compromise safety. In this paper, we propose a Curriculum Reinforcement Learning (CRL) framework integrated with a realistic, data-driven marine simulation environment and a machine learning-based fuel consumption prediction module. The simulation environment is constructed using real-world vessel movement data and enhanced with a Diffusion Model to simulate dynamic maritime conditions. Vessel fuel consumption is estimated using historical operational data and learning-based regression. The surrounding environment is represented as image-based inputs to capture spatial complexity. We design a lightweight, policy-based CRL agent with a comprehensive reward mechanism that considers safety, emissions, timeliness, and goal completion. This framework effectively handles complex tasks progressively while ensuring stable and efficient learning in continuous action spaces. We validate the proposed approach in a sea area of the Indian Ocean, demonstrating its efficacy in enabling sustainable and safe vessel navigation.

The rapid development of multimodal large language models (MLLMs) has highlighted their potential in video understanding. However, challenges remain in long video tasks, particularly in integrating visual features with prompt texts. Existing methods naively store processed video frames in a long-term memory bank, but neglect simple yet effective cross-modal integration. To address this, we introduce the instruction-aware memory construction (IaMC) model for long-term video understanding. By integrating visual and textual information, our model can obtain cross-modal features with robust understanding capabilities. These features are stored in a text-visual memory bank, enabling efficient long-term aggregation without surpassing LLM context or GPU memory limits. Experiments on the LVU dataset demonstrate state-of-the-art performance in video understanding and question answering, showcasing the IaMC model’s effectiveness and setting a new benchmark for long-term video analysis. The source code and trained models will be released publicly.

Automatic infrastructure visual inspection using Unmanned Aerial Vehicles (UAVs) enhances efficiency and safety. However, existing approaches lack context-aware path planning capabilities, often leading to redundant inspections without focus. To address this limitation, this study introduces a novel infrastructure inspection paradigm that integrates 3D coverage path planning (3D-CPP), real-time data interpretation, and inspection information management, to generate and refine 3D-CPP progressively based on recorded information and real-time observation. The proposed paradigm consists of two main components. First, a graph-based information management system named Semantic-PolygonGraph is developed to incorporate static information from Industry Foundation Classes (IFC) alongside dynamically accumulated inspection data. Second, leveraging this structured representation, a progressive 3D-CPP method is proposed to generates an adaptive inspection path that dynamically refines itself based on task requirements, historical records, and real-time observations, prioritizing regions exhibiting superficial damage. To evaluate the effectiveness of the proposed paradigm, this study introduces a data quality assessment metric to quantify the trade-off between inspection cost and data quality. Simulated case studies demonstrate that the proposed approach improves data quality with limited increase of inspection costs, highlighting its potential for long-term infrastructure maintenance.

Digitalisation is transforming various economic sectors, with the digital twin (DT) being a key manifestation for complex systems. While numerous studies focus on sector-specific DTs, few offer comparative analyses across domains. This paper delivers three major contributions: (1) A six-dimensional characterisation framework that systematically captures DT development processes across conceptual (twinning objects, purposes, system architectures) and implementation (data, modelling, services) dimensions; (2) Cross-domain comparative analysis of DTs across five representative domains (agriculture, manufacturing, construction, healthcare, smart cities) using this framework, revealing universal commonalities in DIKW-based intelligence progression and identifying three key differentiators—digitalisation capability, cost-benefit dynamics, and socio-ethical risks—that explain domain-specific variations in DT maturity and adoption; and (3) A unified Digital Twin Platform-as-a-Service (DT-PaaS) solution that standardises common processes, tools, and applications while accommodating domain-specific variations through interoperable data models, reusable modelling libraries, and cross-domain service orchestration. A case study demonstrates that the proposed DT-PaaS framework enables connected DT ecosystems with capabilities for data synchronisation, co-simulation, collaborative learning, and coordinated decision-making across sectors. This research establishes the first systematic cross-domain DT comparison methodology and provides practical pathways for knowledge transfer between mature and emerging DT domains, ultimately supporting more efficient and interoperable digital transformation.

BIM standard development is central to the performance and behavior of BIM model application across transmission, visualization, and information management perspectives. Tremendous effort has been made to ease the implementation of IFC data model in practice. Yet, the complexity of IFC data model hurdles the implementation of the import and export functionality by software vendors. To overcome this, buildingSMART introduced the concept of Model View Definitions to define which parts of an IFC data model need to be implemented for a specific data exchange scenario. With such, the certification of compatibility for software products with the IFC standard is formed. The Model View Definition is use case orientated to determine whether the specific information should be included in an IFC partial model. With the creation of ad-hoc, project-specific Exchange Requirements increasing, associated MVD development requires much more work to incorporate standard development. To resolve this issue, this paper attempts to exploit the potential of information entropy which has proven itself extremely crucial in many other industries in terms of information management, and then integrates it with sandpile simulation to propose a Top-down hierarchy to structure as well as interpret IFC partial model via Model View Definition. The proposed information entropy shifted MVD development approach would manage to unify the MVD development process that enables the reduction on confusion for various end users, specific organization, or project needs. Moreover, to better translate the BIM standard topology into sandpile simulations, a new notion system is proposed. Sandpile simulations are further implemented to prove their applicability, during the simulation, self-organized criticality is identified, and the existence of chaos is observed.

Autonomous bridge visual inspection is a real-world challenge due to various materials, surface coatings, and changing light and weather conditions. Traditional supervised learning relies on massive annotated data to establish a robust model, which requires a time-consuming data acquisition process. This work proposes a few-shot learning (FSL) approach based on improved ProtoNet for damage detection with just a few labeled examples. Feature embedding is achieved through cross-domain transfer learning from ImageNet instead of episodic training. The ProtoNet is improved with embedding normalization to enhance transduction performance based on Euclidean distance and a linear classifier for classification. The approach is explored on a public dataset through different ablation experiments and achieves over 94% mean accuracy for 2-way 5-shot classification via the pre-trained GoogleNet after fine-tuning. Moreover, the proposed fine-tuning methods based on a fully connected layer (FCN) and Hadamard product are demonstrated with better performance than the previous method. Finally, the approach is validated using real bridge inspection images, demonstrating its capability of fast implementation for practical damage inspection with weakly supervised information.

With increasing availability of instrumented infrastructures in built environments, it is necessary to understand how such data will be stored, processed and analysed in a timely manner. Many "smart cities" applications, for instance, identify how data from building sensors can be combined together to support applications such as emergency response, energy management, etc. Enabling sensor data to be transmitted to a Cloud environment for processing provides a number of benefits, such as scalability and elastic provisioning of computational resources - as the total data size may not be known apriori. In this application-based case study, we describe the integration of an in-building sensor network (both for sensing and actuation) with a distributed Cloud environment. Energy optimisation in buildings represents a class of problems that requires significant computational resources and generally is a time consuming process. We describe the use of Cloud computing for efficiently running and deploying Energy Plus simulations with sensor data in order to fulfil a number of energy related objectives for buildings. We describe and evaluate the establishment of such a sensor based application using a Comet Cloud implementation with data collection from a real building pilot. Although our focus is on a single application, the general architecture and analysis carried out can be generalised to other similar scenarios.

To improve the performance of building facility management tools and optimize operational costs, a semantic model is being developed that is synchronized in almost real time with a range of sensors throughout the building. The semantic aspect of the model will be comprised of novel ontological perspectives with identified scopes and restricted domain theories that will simplify the modeling representations to deliver easier knowledge maintenance and continued integrity of the overall system. Specialized ontologies are linked to shared higher level abstract ontologies (to facilitate interoperation), and target a wide range of domains including those to capture physical processes and product models.