When a fire breaks out in a high-rise building,the occlusion of smoke and obstacles results in dearth of crucial information concerning people in distress,thereby creating a challenge in their detection.Given the rest...When a fire breaks out in a high-rise building,the occlusion of smoke and obstacles results in dearth of crucial information concerning people in distress,thereby creating a challenge in their detection.Given the restricted sensing range of a single unmanned aerial vehicle(UAV)cam-era,enhancing the target recognition rate becomes challenging without target information.To tackle this issue,this paper proposes a multi-agent autonomous collaborative detection method for multi-targets in complex fire environments.The objective is to achieve the fusion of multi-angle visual information,effectively increasing the target’s information dimension,and ultimately address-ing the problem of low target recognition rate caused by the lack of target information.The method steps are as follows:first,the you only look once version5(YOLOv5)is used to detect the target in the image;second,the detected targets are tracked to monitor their movements and trajectories;third,the person re-identification(ReID)model is employed to extract the appearance features of targets;finally,by fusing the visual information from multi-angle cameras,the method achieves multi-agent autonomous collaborative detection.The experimental results show that the method effectively combines the visual information from multi-angle cameras,resulting in improved detec-tion efficiency for people in distress.展开更多
Multi-agent reinforcement learning(MARL)has proven its effectiveness in cooperative multi-agent systems(MASs)but still faces issues on the curse of dimensionality and learning efficiency.The main difficulty is caused ...Multi-agent reinforcement learning(MARL)has proven its effectiveness in cooperative multi-agent systems(MASs)but still faces issues on the curse of dimensionality and learning efficiency.The main difficulty is caused by the strong inter-agent coupling nature embedded in an MARL problem,which is yet to be fully exploited in existing algorithms.In this work,we recognize a learning graph characterizing the dependence between individual rewards and individual policies.Then we propose a graph-based reward aggregation(GRA)method,which utilizes the inherent coupling relationship among agents to eliminate redundant information.Specifically,GRA passes information among cooperating agents through graph attention networks to obtain aggregated rewards that contribute to the fitting of the value function,making each agent learn a decentralized executable cooperation policy.In addition,we propose a variant of GRA,named GRA-decen,which achieves decentralized training and decentralized execution(DTDE)when each agent only has access to information of partial agents in the learning process.We conduct experiments in different environments and demonstrate the practicality and scalability of our algorithms.展开更多
Integrated environmental management is important for sustainable development.Under China’s“Three Lines One Permit”(TLOP)policy,three types of management zones—priority protection,critical control,and general contr...Integrated environmental management is important for sustainable development.Under China’s“Three Lines One Permit”(TLOP)policy,three types of management zones—priority protection,critical control,and general control zones—are established based on the ecological red line,the lower-limit line for environmental quality,and the resource use line.Human activities are regulated through a permit system.Integrated and multifactorial protection of soil,plant,hydrological,and atmospheric elements is promoted at the regional level.A follow-up assessment contributes to the improvement of policy implementation and effectiveness.This study demonstrates the achievements of the TLOP policy in Sichuan Province.Results show that(1)276 protection zones have been established under the ecological red line,covering key ecosystems and protected areas to ensure environmental security.Under the lower-limit line,1,626 functional(priority,key,and general control)zones have been designated to regulate air,water,and soil quality,enhancing environmental capacity and pollution control.(2)Through the integration and merging of the three lines,1,128 integrated management zones have been established,including 375,625,and 128 priority protection,critical control,and general control zones,respectively.Each zone has its own list of environmental permits to regulate human activities according to different environmental protection and natural resource development regimes.(3)The design of the follow-up assessment index system was informed by regional primary functions and industrial structure.The index system for provinces and cities is structured around three primary indicators—implementation updating,application,and guarantees—and 15 secondary indicators.The system for critical control zones is structured around environmental access,management,and effectiveness and 14 secondary indicators.A stringent environmental zoning system has been established through the TLOP policy,thereby safeguarding environmental security,promoting harmonious existence between humans and nature,and supporting the vision of Beautiful China.展开更多
This paper presents an adaptive multi-agent coordination(AMAC)strategy suitable for complex scenarios,which only requires information exchange between neighbouring robots.Unlike traditional multi-agent coordination me...This paper presents an adaptive multi-agent coordination(AMAC)strategy suitable for complex scenarios,which only requires information exchange between neighbouring robots.Unlike traditional multi-agent coordination methods that are solved by neural dynamics,the proposed strategy displays greater flexibility,adaptability and scalability.Furthermore,the proposed AMAC strategy is reconstructed as a time-varying complex-valued matrix equation.By introducing a dynamic error function,a fixed-time convergent zeroing neural network(FTCZNN)model is designed for the online solution of the AMAC strategy,with its convergence time upper bound derived theoretically.Finally,the effectiveness and applicability of the coordination control method are demonstrated by numerical simulations and physical experiments.Numerical results indicate that this method can reduce the formation error to the order of 10^(-6)within 1.8 s.展开更多
This paper addresses the synchronization of follower agents’state vectors with that of a leader in high-order nonlinear multi-agent systems.The proposed low-complexity control scheme employs high-gain observers to es...This paper addresses the synchronization of follower agents’state vectors with that of a leader in high-order nonlinear multi-agent systems.The proposed low-complexity control scheme employs high-gain observers to estimate higher-order synchronization errors,enabling the controller to rely solely on relative output measurements.This approach significantly reduces the dependence on full-state information,which is often infeasible or costly in practical engineering applications.An output feedback control strategy is developed to overcome these limitations while ensuring robust and effective synchronization.Simulation results are provided to demonstrate the effectiveness of the proposed approach and validate the theoretical findings.展开更多
The accelerating impacts of climate change,rising temperatures,extreme weather events,and biodiversity loss underscore the urgent need for widespread public awareness.This research explores why climate change awarenes...The accelerating impacts of climate change,rising temperatures,extreme weather events,and biodiversity loss underscore the urgent need for widespread public awareness.This research explores why climate change awareness is not just beneficial but essential for effective environmental stewardship and the long-term health of our planet.The research proffers informed communities,encouraging sustainable practices,and driving policy advocacy,awareness serves as a model for collective action.This call to consciousness challenges individuals,institutions,and nations to recognize their role in shaping a resilient,sustainable future for the Earth.Methodology adopted in this research is a mixed-method design,involving both qualitative and quasi-experimental designs,which engages the use of focus group discussions and oral interviews to explore deeper insights into perceptions,biodiversity loss consciousness,and environmental depletion challenges.Also,applicable under the qualitative method is the secondary data collection mode,namely,reports from IPCC,government policy documents,and existing literature related to the context of the research.The empirical and scientific data analysis was presented from the data collected and was coded and subjected to analysis using a paired samples t-test.The study is grounded on the theory of“Value-Belief-Norm”(VBN)developed by Stern et al.The VBN theory posits that individuals are more likely to engage in pro-environmental behaviour when their values(especially biosphere and altruistic),beliefs(about environmental consequences),and norms(personal responsibility to act)align.The findings of this study underscore the critical role of climate change awareness in fostering environmental and earth stewardship.The paper recommends that Governments of the country(State and federal)should take urgent steps in sensitising the general public on the causes and impact of climate change.展开更多
Multi-Agent Systems(MAS),which consist of multiple interacting agents,are crucial in Cyber-Physical Systems(CPS),because they improve system adaptability,efficiency,and robustness through parallel processing and colla...Multi-Agent Systems(MAS),which consist of multiple interacting agents,are crucial in Cyber-Physical Systems(CPS),because they improve system adaptability,efficiency,and robustness through parallel processing and collaboration.However,most existing unsupervised meta-learning methods are centralized and not suitable for multi-agent systems where data are distributed stored and inaccessible to all agents.Meta-GMVAE,based on Variational Autoencoder(VAE)and set-level variational inference,represents a sophisticated unsupervised meta-learning model that improves generative performance by efficiently learning data representations across various tasks,increasing adaptability and reducing sample requirements.Inspired by these advancements,we propose a novel Distributed Unsupervised Meta-Learning(DUML)framework based on Meta-GMVAE and a fusion strategy.Furthermore,we present a DUML algorithm based on Gaussian Mixture Model(DUMLGMM),where the parameters of the Gaussian-mixture are solved by an Expectation-Maximization algorithm.Simulations on Omniglot and Mini Image Net datasets show that DUMLGMM can achieve the performance of the corresponding centralized algorithm and outperform non-cooperative algorithm.展开更多
This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consen...This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consensus under average dwell time switching.Then sufficient conditions are derived to guarantee the positive consensus.The gain matrices of the control protocol are described using a matrix decomposition approach and the corresponding computational complexity is reduced by resorting to linear programming and co-positive Lyapunov functions.Finally,two numerical examples are provided to illustrate the results obtained.展开更多
In recent years,researchers have leveraged single-agent reinforcement learning to boost educational outcomes and deliver personalized interventions;yet this paradigm provides no capacity for inter-agent interaction.Mu...In recent years,researchers have leveraged single-agent reinforcement learning to boost educational outcomes and deliver personalized interventions;yet this paradigm provides no capacity for inter-agent interaction.Multi-agent reinforcement learning(MARL)overcomes this limitation by allowing several agents to learn simultaneously within a shared environment,each choosing actions that maximize its own or the group's rewards.By explicitly modeling and exploiting agent-to-agent dynamics,MARL can align those interactions with pedagogical goals such as peer tutoring,collaborative problem-solving,or gamified competition,thus opening richer avenues for adaptive and socially informed learning experiences.This survey investigates the impact of MARL on educational outcomes by examining evidence of its effectiveness in enhancing learner performance,engagement,equity,and reducing teacher workload compared to single agent or traditional approaches.It explores the educational domains and pedagogical problems addressed by MARL,identifies the algorithmic families used,and analyzes their influence on learning.The review also assesses experimental settings and evaluation metrics to determine ecological validity,and outlines current challenges and future research directions in applying MARL to education.展开更多
With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier...With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier heterogeneous architecture composed of mobile devices,unmanned aerial vehicles(UAVs),and macro base stations(BSs).This scenario typically faces fast channel fading,dynamic computational loads,and energy constraints,whereas classical queuing-theoretic or convex-optimization approaches struggle to yield robust solutions in highly dynamic settings.To address this issue,we formulate a multi-agent Markov decision process(MDP)for an air-ground-fused MEC system,unify link selection,bandwidth/power allocation,and task offloading into a continuous action space and propose a joint scheduling strategy that is based on an improved MATD3 algorithm.The improvements include Alternating Layer Normalization(ALN)in the actor to suppress gradient variance,Residual Orthogonalization(RO)in the critic to reduce the correlation between the twin Q-value estimates,and a dynamic-temperature reward to enable adaptive trade-offs during training.On a multi-user,dual-link simulation platform,we conduct ablation and baseline comparisons.The results reveal that the proposed method has better convergence and stability.Compared with MADDPG,TD3,and DSAC,our algorithm achieves more robust performance across key metrics.展开更多
To maximize the profits of power grid operators(GOs),load aggregators(LAs)and electricity customers(ECs),this paper proposes a hierarchical demand response(HDR)framework that considers competing interaction based on m...To maximize the profits of power grid operators(GOs),load aggregators(LAs)and electricity customers(ECs),this paper proposes a hierarchical demand response(HDR)framework that considers competing interaction based on multiagent deep deterministic policy gradient(MaDDPG).The ECs are divided into conventional ECs and the electric vehicles(EVs)which are managed by ECs agent(ECA)and EV agent(EVA)to exploit the flexibility of the HDR framework.Thus,the HDR is a tri-layer model determined by five types of agents engaging in competing interaction to maximize their own profits.To address the limitations of mathematical expression and participation scale in the Stackelberg game within the HDR model,a dynamic interaction mechanism is adopted.Moreover,to tackle the HDR involving various entities,the MaDDPG develops multiple agents to simulation the dynamic competing interactions between each subject as well as solve the problem of continuous action control.Furthermore,MaDDPG adopts soft target update and priority experience replay method to ensure stable and effective training,and makes the exploration strategy comprehensive by using exploration noise.Simulation studies are conducted to verify the performance of the MaDDPG with dynamic interaction mechanism in dealing with multilayer multi-agent continuous action control,compared to the double deep Q network(DDQN),deep Q network(DQN)and dueling DQN.Additionally,comparisons among the proposed HDR with the price based DR(PBDR)and incentive based DR(IBDR)are analyzed to investigate the flexibility of the HDR.展开更多
This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external di...This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external disturbances.Under the directed topology conditions,an observer-based finite-time control strategy based on adaptive backstepping and is proposed,in which a neural network-based state observer is employed to approximate the unmeasurable system state variables.To address the complexity explosion problem associated with the backstepping method,a finite-time command filter is incorporated,with error compensation signals designed to mitigate the filter-induced errors.Additionally,the Butterworth low-pass filter is introduced to avoid the algebraic ring problem in the design of the controller.The finite-time stability of the closed-loop system is rigorously analyzed with the finite-time Lyapunov stability criterion,validating that all closed-loop signals of the system remain bounded within a finite time.Finally,the effectiveness of the proposed control strategy is verified through a simulation example.展开更多
Starting from the foundational static traits underlying the growth and development of flue-cured tobacco, this research conducts a systematic examination of the phenomena and theoretical principles associated with env...Starting from the foundational static traits underlying the growth and development of flue-cured tobacco, this research conducts a systematic examination of the phenomena and theoretical principles associated with environment-driven adaptive changes during its cultivation. It was found that environmental variables-including temperature, light, and moisture-elicit directional shifts in static traits ( e.g. , chemical composition, morphological architecture, and leaf tissue structure) toward enhanced environmental adaptation, characterized by graduality, juvenility, similarity, and correlativity. Upon alterations in ambient conditions, flue-cured tobacco modulates its static traits through integrated physical, chemical, and biological-genetic mechanisms, aiming to optimize resource utilization, mitigate environmental constraints, and preserve internal homeostasis alongside metabolic balance. The investigation further reveals that the adaptive scope of flue-cured tobacco to field environments is malleable and can be extended and elevated via adaptive conditioning commencing at the juvenile stage. In addition, the adaptive alignment between static traits and environmental parameters exerts a substantial impact on the plant s growth dynamics, yield performance, and quality attributes. Beyond its relevance to flue-cured tobacco, the proposed theory offers a meaningful framework for elucidating the pervasive adaptive strategies employed by plants and broader biological systems in response to environmental contingencies.展开更多
Addressing optimal confrontation methods in multi-agent attack-defense scenarios is a complex challenge.Multi-Agent Reinforcement Learning(MARL)provides an effective framework for tackling sequential decision-making p...Addressing optimal confrontation methods in multi-agent attack-defense scenarios is a complex challenge.Multi-Agent Reinforcement Learning(MARL)provides an effective framework for tackling sequential decision-making problems,significantly enhancing swarm intelligence in maneuvering.However,applying MARL to unmanned swarms presents two primary challenges.First,defensive agents must balance autonomy with collaboration under limited perception while coordinating against adversaries.Second,current algorithms aim to maximize global or individual rewards,making them sensitive to fluctuations in enemy strategies and environmental changes,especially when rewards are sparse.To tackle these issues,we propose an algorithm of MultiAgent Reinforcement Learning with Layered Autonomy and Collaboration(MARL-LAC)for collaborative confrontations.This algorithm integrates dual twin Critics to mitigate the high variance associated with policy gradients.Furthermore,MARL-LAC employs layered autonomy and collaboration to address multi-objective problems,specifically learning a global reward function for the swarm alongside local reward functions for individual defensive agents.Experimental results demonstrate that MARL-LAC enhances decision-making and collaborative behaviors among agents,outperforming the existing algorithms and emphasizing the importance of layered autonomy and collaboration in multi-agent systems.The observed adversarial behaviors demonstrate that agents using MARL-LAC effectively maintain cohesive formations that conceal their intentions by confusing the offensive agent while successfully encircling the target.展开更多
The coastal regions of southern China experience the country's most frequent convective weather.Accurately representing the low-level upstream atmospheric state over the data-sparse South China Sea(SCS)is crucial ...The coastal regions of southern China experience the country's most frequent convective weather.Accurately representing the low-level upstream atmospheric state over the data-sparse South China Sea(SCS)is crucial for reliable convection predictions in numerical models.Utilizing 10 years of radiosonde observations launched over the SCS,this study presents the upstream offshore convective environments and evaluates the global model data performance including NCEP FNL,ERA5,CRA-40,JRA-3Q,and MERRA-2.Results show that thermodynamic state variables such as temperature and humidity exhibit greater biases than kinetic variables,particularly at low levels.Deeper-layer parameters exhibit smaller uncertainties,especially wind-related variables,while moisture-related parameters have the largest uncertainties,compared to shallower-layer parameters.All model data tend to underestimate the conditional instability and equilibrium level,while overestimating the condensation level,storm relative helicity(SRH),with minimal bias in lapse rate,convective inhibition,vertical wind shear(VWS),and mean winds.These biases primarily arise from the model data's underestimation of temperature and moisture below 700 hPa and lower wind speeds below 950 hPa.Among the global models,CRA-40 performs best in dynamic parameters,with highest correlation and lowest mean absolute error in low-level winds,SRH,VWS,and mean winds.ERA5 excels in thermodynamic parameters.Additional convective-permitting numerical experiments indicate that minor initial condition errors over the upstream ocean significantly affect coastal rainfall production.The rainfall production on windward coasts is most sensitive to the low-level air temperature errors during nocturnal hours,while the rainfall over the PRD is most sensitive to the low-level wind errors.展开更多
In the context of the revolution in new technologies,a key question is whether the rapid growth of the digital economy,driven by digital technologies,has improved regional innovation performance.Using inter-provincial...In the context of the revolution in new technologies,a key question is whether the rapid growth of the digital economy,driven by digital technologies,has improved regional innovation performance.Using inter-provincial panel data from China(2012–2022)and adopting a business environment perspective,this study applies a Panel Extended Regression Model(PERM),a Panel Simultaneous Equation Model(PSEM),and a Tobit-IV model to analyze how the development of the digital economy influences regional innovation.The results reveal a pronounced U-shaped relationship between the digital economy and the regional innovation performance at the provincial level in China,with the business environment serving as a significant mediator in this relationship.Moreover,regional innovation performance in China exhibits a“ratchet effect,”with the impact of the digital economy varying markedly across regions.While the eastern and western regions have entered an upward phase,whereby the digital economy boosts innovation,the central region displays a weaker effect.Further analysis indicates that the synergy between the business environment and the digital economy in driving innovation remains suboptimal.These findings were supported by robust checks.This study offers theoretical insights and empirical evidence that support the coordinated development of digital government and the digital factor market,as well as business environment reforms that are in alignment with the innovation demands of the digital era.展开更多
The development of infrared engineering technologies for extreme environments remains a formidable challenge due to the inherent trade-offs among optical performance,thermal stability,and mechanical integrity in therm...The development of infrared engineering technologies for extreme environments remains a formidable challenge due to the inherent trade-offs among optical performance,thermal stability,and mechanical integrity in thermal photonic metamaterials(TPMs).This work introduces a novel multi-obj ective design framework and demonstrates the design,fabrication,and validation of a TPM operating under extreme temperatures up to 1873 K.We have established a holistic design framework integrating temperaturedependent neural network and Pareto multi-obj ective optimization to co-optimize spectral response,component light-weighting,and structural efficiency.The framework achieves 100 times faster computation than genetic algorithms.The performance of the designed TPM was evaluated under various atmospheric models and detection distances.The TPM achieved a peak radiance suppression efficiency of 82%and a maximum attenuation of-7.4 dB at 1200-1500 K.Experimentally,we fabricated an all-dielectric TPM using a refractory TiO_(2)/BeO multilayer stack with only 5 layers and 2um total thickness.The optimized structure shows high reflectivity(0.62 at 3-5 um;0.48 at 8-14μm)for radiative suppression and high emissivity(0.87 at 5-8μm)for radiative cooling.The TPM withstands 1873 K for 12 h in air with less than 3%spectral drift,retaining excellent mechanical properties.On high-temperature components,it achieves 40-50%radiative suppression and 40-60 K(~10.1 kW m^(-2))radiative cooling at 1100 K,endures over 20 times thermal shock cycles(>150 K s^(-1),700-1500 K),and maintains stable performance over 5 cycles,with 78%visible and 98%microwave transmittance.This work establishes a new paradigm in the design and application of photonic materials for extreme environments.展开更多
Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal ...Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal alignment,temporal consistency,and robust handling of noisy or incomplete inputs across multiple modalities.We propose Multi Agent-Chain of Thought(CoT),a novel multi-agent chain-of-thought reasoning framework where specialized agents for text,vision,and speech modalities collaboratively construct shared reasoning traces through inter-agent message passing and consensus voting mechanisms.Our architecture incorporates self-reflection modules,conflict resolution protocols,and dynamic rationale alignment to enhance consistency,factual accuracy,and user engagement.The framework employs a hierarchical attention mechanism with cross-modal fusion and implements adaptive reasoning depth based on dialogue complexity.Comprehensive evaluations on Situated Interactive Multi-Modal Conversations(SIMMC)2.0,VisDial v1.0,and newly introduced challenging scenarios demonstrate statistically significant improvements in grounding accuracy(p<0.01),chain-of-thought interpretability,and robustness to adversarial inputs compared to state-of-the-art monolithic transformer baselines and existing multi-agent approaches.展开更多
Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posi...Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posing a significant challenge for accurate low-light object detection.Existing methods struggle to preserve discriminative features while maintaining semantic consistency between low-light and normal-light images.For this purpose,this study proposes a DL-YOLO model specially tailored for low-light detection.To mitigate target feature attenuation introduced by repeated downsampling,we design aMulti-Scale FeatureConvolution(MSF-Conv)module that captures rich,multi-level details via multi-scale feature learning,thereby reducing model complexity and computational cost.For feature fusion,we integrated the C3k2-DWRmodule by embedding the Dilation-wise Residual(DWR)mechanism into the 2-core optimized Cross Stage Partial(C3)framework,achieving efficient feature integration.In addition,we replace conventional localization losses with WIoU(Weighted Intersection over Union),which dynamically adjusts gradient gain according to sample quality,thereby improving localization robustness and precision.Experiments on the ExDark dataset demonstrate that DL-YOLO delivers strong low-light detection performance.The relevant code is published at https://github.com/cym0997/DL-YOLO.展开更多
In fire rescue scenarios,traditional manual operations are highly dangerous,as dense smoke,low visibility,extreme heat,and toxic gases not only hinder rescue efficiency but also endanger firefighters’safety.Although ...In fire rescue scenarios,traditional manual operations are highly dangerous,as dense smoke,low visibility,extreme heat,and toxic gases not only hinder rescue efficiency but also endanger firefighters’safety.Although intelligent rescue robots can enter hazardous environments in place of humans,smoke poses major challenges for human detection algorithms.These challenges include the attenuation of visible and infrared signals,complex thermal fields,and interference frombackground objects,all ofwhichmake it difficult to accurately identify trapped individuals.To address this problem,we propose VIF-YOLO,a visible–infrared fusion model for real-time human detection in dense smoke environments.The framework introduces a lightweight multimodal fusion(LMF)module based on learnable low-rank representation blocks to end-to-end integrate visible and infrared images,preserving fine details while enhancing salient features.In addition,an efficient multiscale attention(EMA)mechanism is incorporated into the YOLOv10n backbone to improve feature representation under low-light conditions.Extensive experiments on our newly constructedmultimodal smoke human detection(MSHD)dataset demonstrate thatVIF-YOLOachievesmAP50 of 99.5%,precision of 99.2%,and recall of 99.3%,outperforming YOLOv10n by a clear margin.Furthermore,when deployed on the NVIDIA Jetson Xavier NX,VIF-YOLO attains 40.6 FPS with an average inference latency of 24.6 ms,validating its real-time capability on edge-computing platforms.These results confirm that VIF-YOLO provides accurate,robust,and fast detection across complex backgrounds and diverse smoke conditions,ensuring reliable and rapid localization of individuals in need of rescue.展开更多
文摘When a fire breaks out in a high-rise building,the occlusion of smoke and obstacles results in dearth of crucial information concerning people in distress,thereby creating a challenge in their detection.Given the restricted sensing range of a single unmanned aerial vehicle(UAV)cam-era,enhancing the target recognition rate becomes challenging without target information.To tackle this issue,this paper proposes a multi-agent autonomous collaborative detection method for multi-targets in complex fire environments.The objective is to achieve the fusion of multi-angle visual information,effectively increasing the target’s information dimension,and ultimately address-ing the problem of low target recognition rate caused by the lack of target information.The method steps are as follows:first,the you only look once version5(YOLOv5)is used to detect the target in the image;second,the detected targets are tracked to monitor their movements and trajectories;third,the person re-identification(ReID)model is employed to extract the appearance features of targets;finally,by fusing the visual information from multi-angle cameras,the method achieves multi-agent autonomous collaborative detection.The experimental results show that the method effectively combines the visual information from multi-angle cameras,resulting in improved detec-tion efficiency for people in distress.
基金supported in part by the National Natural Science Foundation of China(grants 62203073 and 62573068)the Natural Science Foundation of Chongqing,China(grant CSTB2022NSCQMSX0577)。
文摘Multi-agent reinforcement learning(MARL)has proven its effectiveness in cooperative multi-agent systems(MASs)but still faces issues on the curse of dimensionality and learning efficiency.The main difficulty is caused by the strong inter-agent coupling nature embedded in an MARL problem,which is yet to be fully exploited in existing algorithms.In this work,we recognize a learning graph characterizing the dependence between individual rewards and individual policies.Then we propose a graph-based reward aggregation(GRA)method,which utilizes the inherent coupling relationship among agents to eliminate redundant information.Specifically,GRA passes information among cooperating agents through graph attention networks to obtain aggregated rewards that contribute to the fitting of the value function,making each agent learn a decentralized executable cooperation policy.In addition,we propose a variant of GRA,named GRA-decen,which achieves decentralized training and decentralized execution(DTDE)when each agent only has access to information of partial agents in the learning process.We conduct experiments in different environments and demonstrate the practicality and scalability of our algorithms.
基金supported by the National Natural Science Foundation of China(grant numbers 42171085)and the National Key R&D Program of China(Grant No.2024YFF1307801,2024YFF1307804).
文摘Integrated environmental management is important for sustainable development.Under China’s“Three Lines One Permit”(TLOP)policy,three types of management zones—priority protection,critical control,and general control zones—are established based on the ecological red line,the lower-limit line for environmental quality,and the resource use line.Human activities are regulated through a permit system.Integrated and multifactorial protection of soil,plant,hydrological,and atmospheric elements is promoted at the regional level.A follow-up assessment contributes to the improvement of policy implementation and effectiveness.This study demonstrates the achievements of the TLOP policy in Sichuan Province.Results show that(1)276 protection zones have been established under the ecological red line,covering key ecosystems and protected areas to ensure environmental security.Under the lower-limit line,1,626 functional(priority,key,and general control)zones have been designated to regulate air,water,and soil quality,enhancing environmental capacity and pollution control.(2)Through the integration and merging of the three lines,1,128 integrated management zones have been established,including 375,625,and 128 priority protection,critical control,and general control zones,respectively.Each zone has its own list of environmental permits to regulate human activities according to different environmental protection and natural resource development regimes.(3)The design of the follow-up assessment index system was informed by regional primary functions and industrial structure.The index system for provinces and cities is structured around three primary indicators—implementation updating,application,and guarantees—and 15 secondary indicators.The system for critical control zones is structured around environmental access,management,and effectiveness and 14 secondary indicators.A stringent environmental zoning system has been established through the TLOP policy,thereby safeguarding environmental security,promoting harmonious existence between humans and nature,and supporting the vision of Beautiful China.
基金supported by the National Natural Science Foundation of China under Grants 61962023,61562029 and 62466019.
文摘This paper presents an adaptive multi-agent coordination(AMAC)strategy suitable for complex scenarios,which only requires information exchange between neighbouring robots.Unlike traditional multi-agent coordination methods that are solved by neural dynamics,the proposed strategy displays greater flexibility,adaptability and scalability.Furthermore,the proposed AMAC strategy is reconstructed as a time-varying complex-valued matrix equation.By introducing a dynamic error function,a fixed-time convergent zeroing neural network(FTCZNN)model is designed for the online solution of the AMAC strategy,with its convergence time upper bound derived theoretically.Finally,the effectiveness and applicability of the coordination control method are demonstrated by numerical simulations and physical experiments.Numerical results indicate that this method can reduce the formation error to the order of 10^(-6)within 1.8 s.
文摘This paper addresses the synchronization of follower agents’state vectors with that of a leader in high-order nonlinear multi-agent systems.The proposed low-complexity control scheme employs high-gain observers to estimate higher-order synchronization errors,enabling the controller to rely solely on relative output measurements.This approach significantly reduces the dependence on full-state information,which is often infeasible or costly in practical engineering applications.An output feedback control strategy is developed to overcome these limitations while ensuring robust and effective synchronization.Simulation results are provided to demonstrate the effectiveness of the proposed approach and validate the theoretical findings.
文摘The accelerating impacts of climate change,rising temperatures,extreme weather events,and biodiversity loss underscore the urgent need for widespread public awareness.This research explores why climate change awareness is not just beneficial but essential for effective environmental stewardship and the long-term health of our planet.The research proffers informed communities,encouraging sustainable practices,and driving policy advocacy,awareness serves as a model for collective action.This call to consciousness challenges individuals,institutions,and nations to recognize their role in shaping a resilient,sustainable future for the Earth.Methodology adopted in this research is a mixed-method design,involving both qualitative and quasi-experimental designs,which engages the use of focus group discussions and oral interviews to explore deeper insights into perceptions,biodiversity loss consciousness,and environmental depletion challenges.Also,applicable under the qualitative method is the secondary data collection mode,namely,reports from IPCC,government policy documents,and existing literature related to the context of the research.The empirical and scientific data analysis was presented from the data collected and was coded and subjected to analysis using a paired samples t-test.The study is grounded on the theory of“Value-Belief-Norm”(VBN)developed by Stern et al.The VBN theory posits that individuals are more likely to engage in pro-environmental behaviour when their values(especially biosphere and altruistic),beliefs(about environmental consequences),and norms(personal responsibility to act)align.The findings of this study underscore the critical role of climate change awareness in fostering environmental and earth stewardship.The paper recommends that Governments of the country(State and federal)should take urgent steps in sensitising the general public on the causes and impact of climate change.
基金supported by the National Natural Science Foundation of China Youth Fund(No.62101579)。
文摘Multi-Agent Systems(MAS),which consist of multiple interacting agents,are crucial in Cyber-Physical Systems(CPS),because they improve system adaptability,efficiency,and robustness through parallel processing and collaboration.However,most existing unsupervised meta-learning methods are centralized and not suitable for multi-agent systems where data are distributed stored and inaccessible to all agents.Meta-GMVAE,based on Variational Autoencoder(VAE)and set-level variational inference,represents a sophisticated unsupervised meta-learning model that improves generative performance by efficiently learning data representations across various tasks,increasing adaptability and reducing sample requirements.Inspired by these advancements,we propose a novel Distributed Unsupervised Meta-Learning(DUML)framework based on Meta-GMVAE and a fusion strategy.Furthermore,we present a DUML algorithm based on Gaussian Mixture Model(DUMLGMM),where the parameters of the Gaussian-mixture are solved by an Expectation-Maximization algorithm.Simulations on Omniglot and Mini Image Net datasets show that DUMLGMM can achieve the performance of the corresponding centralized algorithm and outperform non-cooperative algorithm.
基金supported by the National Natural Science Foundation of China(62463007,62463005)the Natural Science Foundation of Hainan Province(625RC710,625MS047)+1 种基金the System Control and Information Processing Education Ministry Key Laboratory Open Funding,China(Scip20240119)the Science Research Funding of Hainan University,China(KYQD(ZR)22180,KYQD(ZR)23180).
文摘This paper focuses on the leader-following positive consensus problems of heterogeneous switched multi-agent systems.First,a state-feedback controller with dynamic compensation is introduced to achieve positive consensus under average dwell time switching.Then sufficient conditions are derived to guarantee the positive consensus.The gain matrices of the control protocol are described using a matrix decomposition approach and the corresponding computational complexity is reduced by resorting to linear programming and co-positive Lyapunov functions.Finally,two numerical examples are provided to illustrate the results obtained.
文摘In recent years,researchers have leveraged single-agent reinforcement learning to boost educational outcomes and deliver personalized interventions;yet this paradigm provides no capacity for inter-agent interaction.Multi-agent reinforcement learning(MARL)overcomes this limitation by allowing several agents to learn simultaneously within a shared environment,each choosing actions that maximize its own or the group's rewards.By explicitly modeling and exploiting agent-to-agent dynamics,MARL can align those interactions with pedagogical goals such as peer tutoring,collaborative problem-solving,or gamified competition,thus opening richer avenues for adaptive and socially informed learning experiences.This survey investigates the impact of MARL on educational outcomes by examining evidence of its effectiveness in enhancing learner performance,engagement,equity,and reducing teacher workload compared to single agent or traditional approaches.It explores the educational domains and pedagogical problems addressed by MARL,identifies the algorithmic families used,and analyzes their influence on learning.The review also assesses experimental settings and evaluation metrics to determine ecological validity,and outlines current challenges and future research directions in applying MARL to education.
文摘With the advent of sixth-generation mobile communications(6G),space-air-ground integrated networks have become mainstream.This paper focuses on collaborative scheduling for mobile edge computing(MEC)under a three-tier heterogeneous architecture composed of mobile devices,unmanned aerial vehicles(UAVs),and macro base stations(BSs).This scenario typically faces fast channel fading,dynamic computational loads,and energy constraints,whereas classical queuing-theoretic or convex-optimization approaches struggle to yield robust solutions in highly dynamic settings.To address this issue,we formulate a multi-agent Markov decision process(MDP)for an air-ground-fused MEC system,unify link selection,bandwidth/power allocation,and task offloading into a continuous action space and propose a joint scheduling strategy that is based on an improved MATD3 algorithm.The improvements include Alternating Layer Normalization(ALN)in the actor to suppress gradient variance,Residual Orthogonalization(RO)in the critic to reduce the correlation between the twin Q-value estimates,and a dynamic-temperature reward to enable adaptive trade-offs during training.On a multi-user,dual-link simulation platform,we conduct ablation and baseline comparisons.The results reveal that the proposed method has better convergence and stability.Compared with MADDPG,TD3,and DSAC,our algorithm achieves more robust performance across key metrics.
基金supported by the National Natural Science Foundation of China(No.52477097)the GuangDong Basic and Applied Basic Research Foundation(2023A1515240014)the State Key Laboratory of Advanced Electromagnetic Technology(Grant No.AET 2024KF005).
文摘To maximize the profits of power grid operators(GOs),load aggregators(LAs)and electricity customers(ECs),this paper proposes a hierarchical demand response(HDR)framework that considers competing interaction based on multiagent deep deterministic policy gradient(MaDDPG).The ECs are divided into conventional ECs and the electric vehicles(EVs)which are managed by ECs agent(ECA)and EV agent(EVA)to exploit the flexibility of the HDR framework.Thus,the HDR is a tri-layer model determined by five types of agents engaging in competing interaction to maximize their own profits.To address the limitations of mathematical expression and participation scale in the Stackelberg game within the HDR model,a dynamic interaction mechanism is adopted.Moreover,to tackle the HDR involving various entities,the MaDDPG develops multiple agents to simulation the dynamic competing interactions between each subject as well as solve the problem of continuous action control.Furthermore,MaDDPG adopts soft target update and priority experience replay method to ensure stable and effective training,and makes the exploration strategy comprehensive by using exploration noise.Simulation studies are conducted to verify the performance of the MaDDPG with dynamic interaction mechanism in dealing with multilayer multi-agent continuous action control,compared to the double deep Q network(DDQN),deep Q network(DQN)and dueling DQN.Additionally,comparisons among the proposed HDR with the price based DR(PBDR)and incentive based DR(IBDR)are analyzed to investigate the flexibility of the HDR.
基金supported in part by the Beijing Natural Science Foundation under Grant 4252050in part by the National Science Fund for Distinguished Young Scholars under Grant 62425304in part by the Basic Science Center Programs of NSFC under Grant 62088101.
文摘This paper investigates the consensus tracking control problem for high order nonlinear multi-agent systems subject to non-affine faults,partial measurable states,uncertain control coefficients,and unknown external disturbances.Under the directed topology conditions,an observer-based finite-time control strategy based on adaptive backstepping and is proposed,in which a neural network-based state observer is employed to approximate the unmeasurable system state variables.To address the complexity explosion problem associated with the backstepping method,a finite-time command filter is incorporated,with error compensation signals designed to mitigate the filter-induced errors.Additionally,the Butterworth low-pass filter is introduced to avoid the algebraic ring problem in the design of the controller.The finite-time stability of the closed-loop system is rigorously analyzed with the finite-time Lyapunov stability criterion,validating that all closed-loop signals of the system remain bounded within a finite time.Finally,the effectiveness of the proposed control strategy is verified through a simulation example.
基金Supported by Changsha Tobacco Company Science and Technology Project(2020-2024A04).
文摘Starting from the foundational static traits underlying the growth and development of flue-cured tobacco, this research conducts a systematic examination of the phenomena and theoretical principles associated with environment-driven adaptive changes during its cultivation. It was found that environmental variables-including temperature, light, and moisture-elicit directional shifts in static traits ( e.g. , chemical composition, morphological architecture, and leaf tissue structure) toward enhanced environmental adaptation, characterized by graduality, juvenility, similarity, and correlativity. Upon alterations in ambient conditions, flue-cured tobacco modulates its static traits through integrated physical, chemical, and biological-genetic mechanisms, aiming to optimize resource utilization, mitigate environmental constraints, and preserve internal homeostasis alongside metabolic balance. The investigation further reveals that the adaptive scope of flue-cured tobacco to field environments is malleable and can be extended and elevated via adaptive conditioning commencing at the juvenile stage. In addition, the adaptive alignment between static traits and environmental parameters exerts a substantial impact on the plant s growth dynamics, yield performance, and quality attributes. Beyond its relevance to flue-cured tobacco, the proposed theory offers a meaningful framework for elucidating the pervasive adaptive strategies employed by plants and broader biological systems in response to environmental contingencies.
基金co-supported by the National Natural Science Foundation of China(Nos.72371052 and 71871042).
文摘Addressing optimal confrontation methods in multi-agent attack-defense scenarios is a complex challenge.Multi-Agent Reinforcement Learning(MARL)provides an effective framework for tackling sequential decision-making problems,significantly enhancing swarm intelligence in maneuvering.However,applying MARL to unmanned swarms presents two primary challenges.First,defensive agents must balance autonomy with collaboration under limited perception while coordinating against adversaries.Second,current algorithms aim to maximize global or individual rewards,making them sensitive to fluctuations in enemy strategies and environmental changes,especially when rewards are sparse.To tackle these issues,we propose an algorithm of MultiAgent Reinforcement Learning with Layered Autonomy and Collaboration(MARL-LAC)for collaborative confrontations.This algorithm integrates dual twin Critics to mitigate the high variance associated with policy gradients.Furthermore,MARL-LAC employs layered autonomy and collaboration to address multi-objective problems,specifically learning a global reward function for the swarm alongside local reward functions for individual defensive agents.Experimental results demonstrate that MARL-LAC enhances decision-making and collaborative behaviors among agents,outperforming the existing algorithms and emphasizing the importance of layered autonomy and collaboration in multi-agent systems.The observed adversarial behaviors demonstrate that agents using MARL-LAC effectively maintain cohesive formations that conceal their intentions by confusing the offensive agent while successfully encircling the target.
基金supported by the National Natural Science Foundation of China(Grant Nos.42030610,42275006,41805035,and 42305001)the Guangdong Basic and Applied Basic Research Foundation(Grant No.2024A1515030210)+2 种基金the Guangdong Provincial Marine Meteorology Science Data Center(Grant No.2024B1212070014)the Open Project of the Xiamen Key Laboratory of Straits Meteorology(Grant Nos.HXQX202304 and 2024KF02)the Key Innovation Team of the China Meteorological Administration(Grant No.CMA2023ZD08)。
文摘The coastal regions of southern China experience the country's most frequent convective weather.Accurately representing the low-level upstream atmospheric state over the data-sparse South China Sea(SCS)is crucial for reliable convection predictions in numerical models.Utilizing 10 years of radiosonde observations launched over the SCS,this study presents the upstream offshore convective environments and evaluates the global model data performance including NCEP FNL,ERA5,CRA-40,JRA-3Q,and MERRA-2.Results show that thermodynamic state variables such as temperature and humidity exhibit greater biases than kinetic variables,particularly at low levels.Deeper-layer parameters exhibit smaller uncertainties,especially wind-related variables,while moisture-related parameters have the largest uncertainties,compared to shallower-layer parameters.All model data tend to underestimate the conditional instability and equilibrium level,while overestimating the condensation level,storm relative helicity(SRH),with minimal bias in lapse rate,convective inhibition,vertical wind shear(VWS),and mean winds.These biases primarily arise from the model data's underestimation of temperature and moisture below 700 hPa and lower wind speeds below 950 hPa.Among the global models,CRA-40 performs best in dynamic parameters,with highest correlation and lowest mean absolute error in low-level winds,SRH,VWS,and mean winds.ERA5 excels in thermodynamic parameters.Additional convective-permitting numerical experiments indicate that minor initial condition errors over the upstream ocean significantly affect coastal rainfall production.The rainfall production on windward coasts is most sensitive to the low-level air temperature errors during nocturnal hours,while the rainfall over the PRD is most sensitive to the low-level wind errors.
基金National Social Science Fund of China(18KXS009)the Sichuan Provincial Soft Science Program(22JDR0261)the Sichuan University“From 0 to 1”Innovation Research Program(2021CXC10)。
文摘In the context of the revolution in new technologies,a key question is whether the rapid growth of the digital economy,driven by digital technologies,has improved regional innovation performance.Using inter-provincial panel data from China(2012–2022)and adopting a business environment perspective,this study applies a Panel Extended Regression Model(PERM),a Panel Simultaneous Equation Model(PSEM),and a Tobit-IV model to analyze how the development of the digital economy influences regional innovation.The results reveal a pronounced U-shaped relationship between the digital economy and the regional innovation performance at the provincial level in China,with the business environment serving as a significant mediator in this relationship.Moreover,regional innovation performance in China exhibits a“ratchet effect,”with the impact of the digital economy varying markedly across regions.While the eastern and western regions have entered an upward phase,whereby the digital economy boosts innovation,the central region displays a weaker effect.Further analysis indicates that the synergy between the business environment and the digital economy in driving innovation remains suboptimal.These findings were supported by robust checks.This study offers theoretical insights and empirical evidence that support the coordinated development of digital government and the digital factor market,as well as business environment reforms that are in alignment with the innovation demands of the digital era.
基金supported by National Key Research and Development Program of China(2024YFA1210500,2023YFB4606105)Fundamental Research Center Projects(52488301)of National Natural Science Foundation of China(NSFC)+1 种基金Key Research Program of Frontier Sciences(ZDBS-LYJSC030)of Chinese Academy of SciencesWestern Light Program(xbzg-zdsys-202402)of Chinese Academy of Sciences。
文摘The development of infrared engineering technologies for extreme environments remains a formidable challenge due to the inherent trade-offs among optical performance,thermal stability,and mechanical integrity in thermal photonic metamaterials(TPMs).This work introduces a novel multi-obj ective design framework and demonstrates the design,fabrication,and validation of a TPM operating under extreme temperatures up to 1873 K.We have established a holistic design framework integrating temperaturedependent neural network and Pareto multi-obj ective optimization to co-optimize spectral response,component light-weighting,and structural efficiency.The framework achieves 100 times faster computation than genetic algorithms.The performance of the designed TPM was evaluated under various atmospheric models and detection distances.The TPM achieved a peak radiance suppression efficiency of 82%and a maximum attenuation of-7.4 dB at 1200-1500 K.Experimentally,we fabricated an all-dielectric TPM using a refractory TiO_(2)/BeO multilayer stack with only 5 layers and 2um total thickness.The optimized structure shows high reflectivity(0.62 at 3-5 um;0.48 at 8-14μm)for radiative suppression and high emissivity(0.87 at 5-8μm)for radiative cooling.The TPM withstands 1873 K for 12 h in air with less than 3%spectral drift,retaining excellent mechanical properties.On high-temperature components,it achieves 40-50%radiative suppression and 40-60 K(~10.1 kW m^(-2))radiative cooling at 1100 K,endures over 20 times thermal shock cycles(>150 K s^(-1),700-1500 K),and maintains stable performance over 5 cycles,with 78%visible and 98%microwave transmittance.This work establishes a new paradigm in the design and application of photonic materials for extreme environments.
文摘Multimodal dialogue systems often fail to maintain coherent reasoning over extended conversations and suffer from hallucination due to limited context modeling capabilities.Current approaches struggle with crossmodal alignment,temporal consistency,and robust handling of noisy or incomplete inputs across multiple modalities.We propose Multi Agent-Chain of Thought(CoT),a novel multi-agent chain-of-thought reasoning framework where specialized agents for text,vision,and speech modalities collaboratively construct shared reasoning traces through inter-agent message passing and consensus voting mechanisms.Our architecture incorporates self-reflection modules,conflict resolution protocols,and dynamic rationale alignment to enhance consistency,factual accuracy,and user engagement.The framework employs a hierarchical attention mechanism with cross-modal fusion and implements adaptive reasoning depth based on dialogue complexity.Comprehensive evaluations on Situated Interactive Multi-Modal Conversations(SIMMC)2.0,VisDial v1.0,and newly introduced challenging scenarios demonstrate statistically significant improvements in grounding accuracy(p<0.01),chain-of-thought interpretability,and robustness to adversarial inputs compared to state-of-the-art monolithic transformer baselines and existing multi-agent approaches.
文摘Driven by rapid advances in deep learning,object detection has been widely adopted across diverse application scenarios.However,in low-light conditions,critical visual cues of target objects are severely degraded,posing a significant challenge for accurate low-light object detection.Existing methods struggle to preserve discriminative features while maintaining semantic consistency between low-light and normal-light images.For this purpose,this study proposes a DL-YOLO model specially tailored for low-light detection.To mitigate target feature attenuation introduced by repeated downsampling,we design aMulti-Scale FeatureConvolution(MSF-Conv)module that captures rich,multi-level details via multi-scale feature learning,thereby reducing model complexity and computational cost.For feature fusion,we integrated the C3k2-DWRmodule by embedding the Dilation-wise Residual(DWR)mechanism into the 2-core optimized Cross Stage Partial(C3)framework,achieving efficient feature integration.In addition,we replace conventional localization losses with WIoU(Weighted Intersection over Union),which dynamically adjusts gradient gain according to sample quality,thereby improving localization robustness and precision.Experiments on the ExDark dataset demonstrate that DL-YOLO delivers strong low-light detection performance.The relevant code is published at https://github.com/cym0997/DL-YOLO.
基金funded by the National Natural Science Foundation of China under Grant 62306128the Leading Innovation Project of Changzhou Science and Technology Bureau underGrant CQ20230072+2 种基金the Basic Science Research Project of Jiangsu Provincial Department of Education under Grant 23KJD520003the Science and Technology Development Plan Project of Jilin Provinceunder Grant 20240101382JCthe National KeyR esearch and Development Program of China under Grant 2023YFF1105102.
文摘In fire rescue scenarios,traditional manual operations are highly dangerous,as dense smoke,low visibility,extreme heat,and toxic gases not only hinder rescue efficiency but also endanger firefighters’safety.Although intelligent rescue robots can enter hazardous environments in place of humans,smoke poses major challenges for human detection algorithms.These challenges include the attenuation of visible and infrared signals,complex thermal fields,and interference frombackground objects,all ofwhichmake it difficult to accurately identify trapped individuals.To address this problem,we propose VIF-YOLO,a visible–infrared fusion model for real-time human detection in dense smoke environments.The framework introduces a lightweight multimodal fusion(LMF)module based on learnable low-rank representation blocks to end-to-end integrate visible and infrared images,preserving fine details while enhancing salient features.In addition,an efficient multiscale attention(EMA)mechanism is incorporated into the YOLOv10n backbone to improve feature representation under low-light conditions.Extensive experiments on our newly constructedmultimodal smoke human detection(MSHD)dataset demonstrate thatVIF-YOLOachievesmAP50 of 99.5%,precision of 99.2%,and recall of 99.3%,outperforming YOLOv10n by a clear margin.Furthermore,when deployed on the NVIDIA Jetson Xavier NX,VIF-YOLO attains 40.6 FPS with an average inference latency of 24.6 ms,validating its real-time capability on edge-computing platforms.These results confirm that VIF-YOLO provides accurate,robust,and fast detection across complex backgrounds and diverse smoke conditions,ensuring reliable and rapid localization of individuals in need of rescue.
