Search Results (1,669)

Vision–Language–Action (VLA) models unify visual perception, natural-language understanding, and action generation within a single foundation model, allowing a robot to follow instructions such as “fold the towel” or “fly to the red building” directly from camera images. Because VLAs inherit world knowledge from internet-scale pre-training, they have become the dominant framework for learning-based manipulation, with bimanual coordination serving as the most demanding testbed: two arms with $7+$ degrees of freedom each must move in concert to fold, assemble, and reorient objects. Unmanned aerial robotics faces a structurally similar challenge: a drone must coordinate thrust, attitude, and increasingly gripper commands from visual observations under strict latency and payload constraints. This review covers 183 contributions spanning 2017–2026 and organized along seven dimensions: VLA architectures, training recipes, action representations, bimanual coordination (2022–2026), unmanned aerial vehicle (UAV) navigation and control (2017–2026), language grounding, and cross-cutting concerns including memory and world models. We show that the coordination strategies, training recipes, and action representations developed for bimanual VLAs transfer to unmanned aerial systems and identify fourteen research directions across both domains. Full article

Rabies is a fatal zoonotic viral disease that continues to pose a significant threat to both animal and public health worldwide. Despite considerable progress in its control across Europe, sporadic outbreaks still occur, particularly in regions where wildlife reservoirs and stray animal populations sustain virus circulation. This study provides one of the first comparative longitudinal analyses integrating European countries and Turkiye rabies surveillance data over a decade (2013–2023). Information on reported outbreaks was obtained from the Animal Disease Information System (ADIS) and the World Animal Health Information System (WAHIS) database. The analysis focused on temporal trends, regional differences, and the distribution of affected animal species. During the study period, a total of 4865 outbreaks were reported in 16 countries. The number of detected outbreaks declined considerably over time, decreasing from 1022 cases in 2013 to 325 cases in 2023, representing an overall reduction of approximately 68%. The temporal trend was not uniform, with periods of decline followed by temporary increases. The highest number of outbreaks was registered in Türkiye, followed by Romania and Poland, indicating pronounced regional disparities. Domestic dogs represented the most frequently affected species, while cases were also recorded in wildlife and domestic cats, confirming the epidemiological importance of both domestic and wild reservoirs. The observed reduction in the number of outbreaks reflects the impact of vaccination programs and coordinated control measures, but may also be influenced by differences in surveillance systems and reporting practices. Nevertheless, the persistence of rabies in several regions indicates that the disease remains an epidemiological concern. Sustained vaccination of domestic animals, continued wildlife immunization, and strengthened surveillance and cross-border cooperation are essential for long-term control and prevention. Full article

Climate change poses accelerating and intensifying threats to cultural heritage worldwide, necessitating urgent and coordinated state-level responses. This study critically examines China’s governance framework for climate adaptation of cultural heritage, identifying a critical policy misalignment: although relevant legal and governance instruments—spanning cultural heritage protection, environmental governance, disaster risk reduction, territorial spatial planning, and climate action systems—are nominally in place, they remain profoundly fragmented in practice, resulting in operational inefficiency that severely constrains effective adaptation. To address this, the paper argues for a fundamental paradigm shift from static preservation to dynamic adaptation. It proposes a reform pathway centered on three pillars: reconceptualizing heritage from static preservation to dynamic adaptation, institutionalizing cross-departmental cooperation, and integrating systemic adaptation tools into planning and decision-making. The ultimate objective is to establish an adaptive governance system capable of responding flexibly to climate impacts through interdisciplinary coordination. This transformation is framed as a critical strategic imperative, essential for ensuring the long-term resilience of cultural heritage and civilizational continuity in a warming world. Full article

End-to-end autonomous driving remains fragile in long-tail scenarios, while incorporating vision-language models (VLMs) introduces substantial deliberation latency that cannot interfere with the real-time planning loop. We present ThinkDrive, a dual-process driving framework designed under explicit real-time queuing constraints. The framework contains four coordinated components. First, a Scene Complexity Estimator regulates System-2 activation through a trigger cool-down mechanism, allowing at most one asynchronous request every $\lceil L_2/\Delta t\rceil$ frames and thereby preventing queue saturation under a System-2 latency of $L_2=565$ ms. Second, a multi-modal System-1 planner generates $K_1=5$ candidate trajectories within 44 ms and is trained with winner-takes-all imitation learning together with explicit score supervision. Third, a two-stage Causal-CoT module uses the VLM to identify risk agents and predict a preferred spatial goal $G_{\mathrm{VLM}}$, after which a single batched scm_rollout selects the safest candidate and extracts its endpoint as a world-coordinate goal anchor ${\mathbf{g}}_{S2}$. Fourth, a Goal-Anchor Replanning module transforms ${\mathbf{g}}_{S2}$ into the current ego frame and selects the candidate whose waypoint at the remaining time horizon is closest to the transformed goal. This design avoids coordinate-space mixing, mitigates bias caused by mismatched temporal horizons, and prevents semantic instability across replanning cycles. On nuPlan test14-hard, ThinkDrive with InternVL2-8B and a 6.8% trigger rate achieves 74.9 PDMs, outperforming AdaThinkDrive at 73.1 while maintaining a nominal latency of 44 ms. Full article

Existing self-supervised methods have achieved significant success in ground-level autonomous driving scenarios, but applying them directly to Unmanned Aerial Vehicle (UAV) videos remains challenging. On the one hand, rapid pose changes in UAVs often lead to oblique-view imaging, making it difficult for conventional methods to handle the perspective distortion in oblique imagery. On the other hand, complex UAV viewpoints may cause depth blurring in low-texture regions. To address these challenges, we propose a lightweight self-supervised monocular depth estimation method for UAV scenarios. By utilizing a Dynamic Direction-Aware Module (DDaM), the network adaptively adjusts the sampling grid to correct distorted features during feature extraction, while enhancing its ability to capture features at different spatial locations. Furthermore, to mitigate the loss of spatial information caused by multiple downsampling operations, we integrate a Coordinate Attention Mechanism into the encoder. This mechanism captures features along two separate spatial axes, preserving the spatial coordinates of object boundaries. Our experiments demonstrate that the synergy between DDaM and the Coordinate Attention Mechanism enables the prediction of more accurate object boundaries and richer local details. To validate the effectiveness and practical applicability of the proposed method, we conduct experiments on both the MidAir synthetic dataset and the UAVid real-world dataset. The results show that, compared with current baseline methods, our approach maintains competitive performance while requiring the fewest parameters. Full article

Urban traffic congestion presents a complex challenge driven by intricate spatial dependencies and non-stationary temporal dynamics. While Multi-Agent Deep Reinforcement Learning has shown promise for Traffic Signal Control, existing approaches often struggle with partial observability and fail to coordinate effectively across large-scale, heterogeneous road networks. In this paper, we propose HydraLight (HYbrid Deep Reinforcement Learning Architecture for Traffic Lights), a novel spatio-temporal framework that integrates Graph Attention Networks and Temporal Transformers. To overcome the localized myopia of standard graph methods, HydraLight introduces a Global Pooling Context module that broadcasts macroscopic, citywide traffic summaries, enabling agents to proactively mitigate systemic gridlock. Furthermore, to facilitate robust multi-scenario training, we introduce a Unified Prioritized Experience Replay (Unified PER) module that normalizes Temporal-Difference errors, preventing task dominance across diverse topologies. Extensive experiments on the RESCO benchmark across five synthetic and real-world networks demonstrate that HydraLight consistently outperforms state-of-the-art baselines (including X-Light and CoSLight).Byreducing traffic congestion, travel delays, and idle waiting times, the proposed framework also contributes to more sustainable urban mobility through improved traffic flow efficiency, lower fuel consumption, and reduced vehicular carbon emissions. Notably, the proposed architecture excels in structurally irregular environments, achieving up to 13.07% reduction in average travel time on complex arterial networks and consistently improving queue stability and waiting-time minimization across both synthetic and real-world RESCO benchmarks compared to state-of-the-art baselines. Full article

Urban navigation in dynamic environments remains a challenging problem for blind and visually impaired users due to the presence of unpredictable obstacles and the limitations of conventional navigation systems, which rely primarily on static map-based information and lack real-time environmental awareness. This paper presents a real-time sensor-driven navigation system based on a multi-agent architecture incorporating a reinforcement-learning navigation policy for assistive mobility in urban environments. The proposed system integrates GPS-based global localization with vision-based perception to enable continuous fusion of global route planning and local obstacle detection. This integration allows the system to dynamically adjust navigation strategies in response to changing environmental conditions. The architecture is designed as a modular multi-agent system comprising agents for perception, navigation, sensor fusion, personalization, safety arbitration, interface management, and system monitoring. The reinforcement learning component formulates local navigation as a sequential decision-making problem, where the navigation policy is trained to balance path efficiency, obstacle avoidance, and safety constraints through interaction with simulated environments. Prototype implementation is developed and evaluated in both simulation and controlled real-world scenarios. Experimental results demonstrate that the proposed system shows improved obstacle avoidance performance and navigation stability under the evaluated conditions while maintaining low-latency responsiveness compared to baseline navigation approaches. The system also exhibits robust behaviour under varying environmental conditions, supporting its potential applicability to assistive navigation tasks in controlled urban environments. The proposed approach contributes to a scalable architecture that integrates a reinforcement-learning navigation policy within a multi-agent coordination framework and real-time sensor perception, providing a foundation for the development of intelligent and deployable assistive navigation systems. Full article

Driven by global energy transformation and the progress of artificial intelligence technology, traditional automotive engineering is undergoing profound changes. Transportation is rapidly advancing toward electrification and intelligence. Against this background, this paper identifies three emerging paradigms for the development of electric vehicles: Heart Revolution, Brain Evolution, and Network Integration. This paper points out that automobiles are evolving from traditional one-way energy consumers to dynamic energy nodes in smart grids. With the support of artificial intelligence technology, the role of automobiles is also shifting from a simple means of transportation to an intelligent mobile terminal. At the same time, this paper focuses on analyzing the application of the integration theory of “Four Networks and Four Flows” in automobile upgrading. The theory does not focus on the optimization of a single node unit but emphasizes a systematic perspective to improve overall performance and support sustainable development. This paper suggests that the development of the automobile industry must be deeply integrated with the humanity world, information world and physical world. By building a five-in-one architecture of “Human–Vehicle–Road–Cloud–Satellite”, the automobile industry could follow a practical pathway toward coordinated development. At the same time, breakthroughs in core technologies such as solid-state batteries and wide-bandgap semiconductors are also imminent. This paper aims to provide a sustainable and high-performance automobile development path and integrate the concept of human-oriented design into it. Meanwhile, China’s new energy vehicle industry is used as a representative context to illustrate its engineering and industrial implementation. Full article

Background: Multisite digital psychiatry trials increasingly rely on complex onboarding and implementation processes at local research sites. While outcome-focused evaluations are common, less attention has been paid to the site-level labor required to operationalize such studies in real-world settings, particularly at smaller or resource-constrained sites. This study addresses this gap by examining hidden implementation labor from a single-site reflexive perspective. Methods: This study adopts a reflexive qualitative case study approach to examine onboarding and implementation processes at a single research site participating in a multisite digital psychiatry trial (ClinicalTrials.gov: NCT04953208). The analysis draws on longitudinal experiential data, supported by site-specific documentation, onboarding timelines, troubleshooting records, device-management materials, data quality assurance activities, and internal communications generated during site coordination and implementation activities. Results: Five interrelated themes were identified: hidden labor and role overload; resource scarcity at small research sites; fragmented remote communication and technical coordination; multi-role professional contexts and competing demands; and the impact of external systemic disruptions. Findings show how administrative, technical, logistical, and coordination tasks were absorbed into individual roles, often exceeding initial role expectations. Despite limited resources, the site achieved high performance through intensified individual effort, masking the true implementation burden. This pattern is conceptualized as a high-performance paradox, in which apparent site efficiency may conceal substantial hidden labor and role compression. Conclusions: This site-level reflexive account highlights the central role of hidden labor in sustaining implementation in multisite digital psychiatry trials. Recognizing and explicitly resourcing implementation work, particularly at small research sites, may improve feasibility, sustainability, and equity across study settings. The study contributes a practice-based methodological perspective on how implementation burden can be identified through reflexive analysis of site-level trial processes. Full article

China is the world’s largest producer of hydrogen, and it has the potential to export renewable hydrogen and its derivatives. Japan has set ambitious targets for developing a hydrogen-based society but is facing cost challenges. There is strong potential for China and Japan to cooperate regarding renewable hydrogen across the value chain. This study evaluates the cooperation opportunities from the 3E perspective (energy security, economics, and the environment). It estimates the renewable hydrogen production potential in both countries, as well as the economics and greenhouse gas (GHG) emissions associated with the production and export of renewable hydrogen from China to Japan using proton exchange membrane (PEM) technology. The renewable hydrogen production potential in China is estimated to be 12.00 Mt/year by 2035 in the base case of this study, providing a strong foundation for exports to Japan. The levelized cost of hydrogen (LCOH) using PEM technology and onshore wind is estimated at 4.27 USD/kg H₂ in China and 11.01 USD/kg H₂ in Japan for projects built in 2025. Even after accounting for liquefaction costs in China, transport costs from China to Japan (Chifeng—Dalian—Kobe) and regasification costs in Japan, renewable hydrogen produced in China remains more cost-effective than that produced in Japan. In terms of GHG emissions, when renewable hydrogen is produced using wind power, and wind power is also used for liquefaction and other electricity-consuming processes, the total emissions within the case study boundary amount to 2.24 kg CO₂-eq/kg H₂, below Japan’s low-carbon hydrogen threshold of 3.4 CO₂-eq/kg H₂. This study also discusses the challenges which are critical to facilitating cooperation, particularly in regards to coordinating standards and certification systems between the two countries. Full article

Wearable inertial measurement units are widely used for human motion analysis due to their portability; however, most IMU-based motion capture systems rely on dense sensor configurations that increase cost, complexity, and usability challenges in real-world applications. To address this limitation, this paper presents a sparse inertial human motion reconstruction framework that uses only five wearable sensors while maintaining real-time performance and biomechanical plausibility. The proposed framework integrates Movella Xsens DOT IMUs with a learning-based inverse kinematics pipeline and a real-time biomechanical digital twin for motion reconstruction and visualization. The evaluation was conducted in two phases: first, a real-time motion streaming system was established to validate sensor alignment, coordinate frame consistency, and end-to-end latency; second, a sparse inference framework was trained using the Virginia Tech Natural Motion Dataset combined with a custom dataset containing hard negative samples. Experimental results show that the system can accurately reconstruct full-body human motion, excluding head movement, with a local Mean Per-Joint Position Error of 5.96 cm using only five sensors. Comparative ablation studies demonstrate that Transformer-based temporal modeling achieves better geometric accuracy and temporal smoothness than recurrent and convolutional baselines, while physics-informed regularization and hard negative mining significantly improve biomechanical consistency and reduce motion jitter. Real-time experiments further demonstrate that the framework operates within interactive latency limits, highlighting its potential for biomechanical digital twin applications. Full article

Background: Extracorporeal cardiopulmonary resuscitation (ECPR) has emerged as a promising strategy for refractory cardiac arrest, enabling the restoration of systemic perfusion when conventional resuscitation fails. However, uncertainties remain regarding patient selection, timing and implementation. Methods: A narrative review of experimental data, clinical studies, randomized trials and international recommendations was performed. Particular emphasis was placed on the interplay between physiological mechanisms, real-world organizational models and decision-making processes. Results: ECPR can restore effective circulation, preserve end organ perfusion and serve as a bridge to definitive etiologic treatment, with the potential to improve survival and neurological outcomes in highly selected patients. However, its effectiveness is strongly dependent on rapid deployment, structured systems of care and multidisciplinary coordination. Significant challenges remain, including in relation to the heterogeneity of protocols, high resource utilization, complications with extracorporeal support and the complexity of post-resuscitation management. Furthermore, ECPR fundamentally alters traditional resuscitation paradigms, introducing ethical dilemmas related to patient selection, prognostication and the allocation of limited resources. Conclusions: ECPR represents a transition from procedure-based resuscitation to system-based extracorporeal support. Its clinical benefit is contingent upon timely implementation within optimized organizational frameworks and integration with definitive treatment pathways. Future research should focus on refining selection criteria, standardizing care pathways and addressing ethical sustainability challenges to ensure appropriate and effective use of this evolving technology. Full article

Background/Objectives: Vaccine development pipelines are forward-looking indicators of public health preparedness, reflecting the capacity to address unmet medical needs and emerging threats. This descriptive analysis aims to characterise the 2025 clinical-stage pipeline of infectious disease vaccines and prophylactic monoclonal antibody candidates developed by Vaccines Europe member companies, and to describe how pipeline characteristics address evolving public health priorities. Methods: A descriptive analysis was conducted using publicly available data compiled in the Vaccines Europe Pipeline Review 2025, with validation by participating companies. Candidates in clinical development or regulatory review were classified using a standardised framework by pathogen/disease, target population, public health priority, and technologies. Results: The Vaccines Europe member company pipeline comprises 91 candidates across clinical development phases, 19% of which are in Phase III and 7% undergoing regulatory review. This pipeline is predominantly targeting respiratory pathogens (75%), with a strong life-course focus (85% evaluated in adults and/or older adults), and sustained activity in bacterial pathogens relevant to antimicrobial resistance. Notably, 41% of candidates were classified as addressing diseases, disease combinations, or indications for which no licenced preventive product exists. This category includes candidates targeting diseases without a preventive solution, as well as novel combination vaccines and therapeutic approaches in areas where individual components or preventive vaccines are already available. This captures vaccines candidates in different stages of development, not necessarily first-in-disease innovation. The pipeline shows broad technological diversity (12 technologies), dominated by RNA approaches and multivalent candidates, with growing focus on climate-sensitive, zoonotic, and pandemic-prone pathogens. Conclusions: Within the pipeline of Vaccines Europe member companies, this analysis describes development activity oriented toward broader prevention, platform-based approaches, and preparedness-relevant targets. As a structured and recurring annual assessment, the Vaccines Europe Pipeline Review supports horizon scanning and evidence-based dialogue between industry and vaccine ecosystem stakeholders. In order to maximise the impact of vaccine development pipelines to public health, predictable investment, streamlined trial and regulatory pathways, strong surveillance, and real-world data systems, coordinated decision-making is required to enable timely and equitable access, and complementary incentive and procurement reforms. Full article

Smart building services are increasingly realised through combinations of sensors, actuators, communication infrastructures, software platforms, analytics, and artificial intelligence-based functions. These configurations enable adaptive control, real-time monitoring, contextual automation, predictive support, user interaction, and cross-domain coordination across heating, ventilation, air conditioning, lighting, energy management, security and access control, water management, and user-centric comfort services. At the same time, the European Union Smart Readiness Indicator provides a formal basis for assessing building smartness through technical domains, service functionalities, and multidimensional impact criteria. A systematic basis for translating real-world descriptions of smart building services and their enabling technology stacks into Smart Readiness Indicator-aligned assessment inputs remains underdeveloped. A PRISMA ScR informed review was conducted to identify principal smart building service domains, synthesise their core functionalities, and reconstruct the digital technologies through which these functionalities are realised. The synthesis shows that heating, ventilation, and air conditioning and lighting provide comparatively direct translation pathways to formal Smart Readiness Indicator domains, while energy management operates mainly as a supervisory and cross-domain layer. Security and access control, water management, and several user-centric services contribute meaningfully to building smartness but often show partial or extended formal correspondence. Monitoring and control emerge as a central cross-cutting layer because many higher-order smart building capabilities are expressed through visibility, supervision, orchestration, and digital representation. Building on this review, a methodological framework is established for translating smart building services into Smart Readiness Indicator-aligned assessments. The procedure uses the smart building service instance as the unit of analysis and links service identification, functionality formulation, technology stack reconstruction, formal domain correspondence, impact profiling, maturity classification, and building-level aggregation. This enables heterogeneous service descriptions to be converted into structured readiness profiles while preserving the distinction between operational functionality, enabling technology, formal assessment correspondence, and multidimensional impact contribution. Application of the framework to the IoT Building Cloud platform shows that a substantial share of smart building capability may derive from supervisory digital infrastructure rather than from isolated end-use control alone. The resulting readiness profile is characterised by strong representation in monitoring and control, information to occupants and operators, and maintenance awareness, together with more selective contributions to indoor environmental control and limited flexibility-related capability. The proposed framework supports Smart Readiness Indicator-aligned pre-assessment, comparative analysis, design stage reasoning, and digital tool development by providing a transparent bridge between smart building service descriptions and formal assessment-oriented interpretation. Full article

Handover optimization is essential for seamless connectivity in 5G and beyond networks. Existing approaches present fundamental challenges of centralized solutions achieving coordination and accuracy but creating privacy risks under the General Data Protection Regulation (GDPR), while distributed privacy-preserving approaches protect user data but lack the network-wide visibility necessary for optimal mobility decisions. This systematic review synthesizes 49 peer-reviewed studies published between 2010 and 2025, identified through a PRISMA-compliant search across IEEE Xplore, ScienceDirect, SpringerLink, MDPI, ACM Digital Library, and Google Scholar. Eligible studies addressed cellular handover or mobility management using traditional signal-based, Machine Learning, Federated Learning, Software-Defined Networking strategies, and reported quantitative performance metrics. A structured quality assessment evaluated methodological rigor, dataset validation, benchmarking practices, handover-specific metrics, and scalability. Synthesis evidence shows that existing approaches do not simultaneously satisfy critical requirements for next-generation mobility management of accuracy, privacy, scalability, and real-time network-wide coordination. Machine learning achieves high accuracy (up to 97%) but depends on centralized data; Reinforcement Learning supports real-time adaptation but incurs high computational costs; federated learning preserve privacy but suffers from limited global coordination; and software-defined networking enables centralized control but requires continuous transmission of raw data. Evidence quality is further limited to simulation-based assessments and limited real-world datasets. Overall, the reviews identify a clear evolution from reactive threshold-based methods towards proactive prediction and highlights the need for unified, privacy-preserving and globally coordinated handover frameworks. The findings point toward integrating federated learning with Software-Defined Mobile Networking as promising architectural direction for 6G mobility management. Full article