Hydrogen energy is a crucial support for China’s low-carbon energy transition.With the large-scale integration of renewable energy,the combination of hydrogen and integrated energy systems has become one of the most ...Hydrogen energy is a crucial support for China’s low-carbon energy transition.With the large-scale integration of renewable energy,the combination of hydrogen and integrated energy systems has become one of the most promising directions of development.This paper proposes an optimized schedulingmodel for a hydrogen-coupled electro-heat-gas integrated energy system(HCEHG-IES)using generative adversarial imitation learning(GAIL).The model aims to enhance renewable-energy absorption,reduce carbon emissions,and improve grid-regulation flexibility.First,the optimal scheduling problem of HCEHG-IES under uncertainty is modeled as a Markov decision process(MDP).To overcome the limitations of conventional deep reinforcement learning algorithms—including long optimization time,slow convergence,and subjective reward design—this study augments the PPO algorithm by incorporating a discriminator network and expert data.The newly developed algorithm,termed GAIL,enables the agent to perform imitation learning from expert data.Based on this model,dynamic scheduling decisions are made in continuous state and action spaces,generating optimal energy-allocation and management schemes.Simulation results indicate that,compared with traditional reinforcement-learning algorithms,the proposed algorithmoffers better economic performance.Guided by expert data,the agent avoids blind optimization,shortens the offline training time,and improves convergence performance.In the online phase,the algorithm enables flexible energy utilization,thereby promoting renewable-energy absorption and reducing carbon emissions.展开更多
As a green sustainable alternative technology,synthesizing nitrate by electrocatalytic nitrogen oxidation reaction(NOR)can replace the traditional energyintensive Ostwald process.But low nitrogen fixation yields and p...As a green sustainable alternative technology,synthesizing nitrate by electrocatalytic nitrogen oxidation reaction(NOR)can replace the traditional energyintensive Ostwald process.But low nitrogen fixation yields and poor selectivity due to the high bond energy of the N≡N bond and competition from the oxygen evolution reaction in the electrolyte restrict its application.On the other hand,two-dimensional(2D)PdS_(2)as a member in the family of group-10 novel transition metal dichalcogenides(NTMDs)presents the interesting optical and electronic properties due to its novel folded pentagonal structure,but few researches involve to its fabrication and application.Herein,unique imitating growth feature for PdS_(2)on different 2D substrates has been firstly discovered for constructing 2D/2D heterostructures by interface engineering.Due to the different exposed chemical groups on the substrates,PdS_(2)grows as the imitation to the morphologies of the substrates and presents different thickness,size,shape and the degree of oxidation,resulting in the significant difference in the NOR activity and stability of the obtained composite catalysts.Especially,the thin and small PdS_(2)nanoplates with more defects can be obtained by decorating poly(1-vinyl-3-ethylimidazolium bromide)on the 2D substrate,easily oxidized during the preparation process,resulting in the in situ generation of SO_(4)^(2−),which plays a crucial role in reducing the activation energy of the NOR process,leading to improved efficiency for nitrate production,verified by theoretical calculation.This research provides valuable insights for the development of novel electrocatalysts based on NTMDs for NOR and highlights the importance of interface engineering in enhancing catalytic performance.展开更多
The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location re...The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location reidentification and correlation attacks.To address these challenges,privacy-preserving trajectory generation methods are critical for applications relying on sensitive location data.This paper introduces DPIL-Traj,an advanced framework designed to generate synthetic trajectories while achieving a superior balance between data utility and privacy preservation.Firstly,the framework incorporates Differential Privacy Clustering,which anonymizes trajectory data by applying differential privacy techniques that add noise,ensuring the protection of sensitive user information.Secondly,Imitation Learning is used to replicate decision-making behaviors observed in real-world trajectories.By learning from expert trajectories,this component generates synthetic data that closely mimics real-world decision-making processes while optimizing the quality of the generated trajectories.Finally,Markov-based Trajectory Generation is employed to capture and maintain the inherent temporal dynamics of movement patterns.Extensive experiments conducted on the GeoLife trajectory dataset show that DPIL-Traj improves utility performance by an average of 19.85%,and in terms of privacy performance by an average of 12.51%,compared to state-of-the-art approaches.Ablation studies further reveal that DP clustering effectively safeguards privacy,imitation learning enhances utility under noise,and the Markov module strengthens temporal coherence.展开更多
Unmanned aerial vehicles(UAVs)face the challenge of autonomous obstacle avoidance in complex,multi-obstacle environments.Behavior cloning offers a promising approach to rapidly acquire a learning policy from limited e...Unmanned aerial vehicles(UAVs)face the challenge of autonomous obstacle avoidance in complex,multi-obstacle environments.Behavior cloning offers a promising approach to rapidly acquire a learning policy from limited expert demonstrations.However,pure imitation learning inherently suffers from poor exploration and limited generalization,typically necessitating extensive datasets to train competent student policies.We utilize a cross-modal variational autoencoder(CM-VAE)to extract compact features from raw visual inputs and UAV states,which then feed into a policy network.We evaluated our approach in a simulated environment featuring a challenging circular trajectory with eight gate obstacles.The results demonstrate that the policy trained with pure behavior cloning consistently failed.In stark contrast,our DAgger-augmented behavior cloning method successfully traversed all gates without collision.Our findings confirm that DAgger effectively mitigates the shortcomings of behavior cloning,enabling the creation of reliable and sample-efficient navigation policies for UAVs.展开更多
Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional ...Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.展开更多
It is of vital importance for modern college English teaching to properly construct an interactive multimedia-internet-based teaching system, the structure of which is clearly elaborated in this paper. An IMITS usuall...It is of vital importance for modern college English teaching to properly construct an interactive multimedia-internet-based teaching system, the structure of which is clearly elaborated in this paper. An IMITS usually consists of hardware, software, teaching and management. At the end of this paper, a conclusion is made that only when all the four parts of IMITS are construct ed such as is demonstrated, can the IMITS exert its full effects in college English teaching.展开更多
Environmental comprehensive management system, called “the bionic community”, can be established in imitation of biome, which can transform the wastes generated in a certain field into the raw materials of other fie...Environmental comprehensive management system, called “the bionic community”, can be established in imitation of biome, which can transform the wastes generated in a certain field into the raw materials of other field. The establishment of the bionic community includes two aspects, i.e., the matching technique and the management system. The main matching technique is the preparation of composite materials made of various wastes. This new kind of material can be divided into four types: polymer matrix, silicate matrix, metal matrix and carbon matrix(or ceramic matrix). The environmental comprehensive management system is formed by organizing a trans-trades joint-management business entity with the products of composite material made of wastes at the core.展开更多
Owing to the diversity of consumer’s demand, the traditional Sichuan brocade products have to be innovated in order to survive in a competitive environment. Since the construction of traditional Sichuan brocade was c...Owing to the diversity of consumer’s demand, the traditional Sichuan brocade products have to be innovated in order to survive in a competitive environment. Since the construction of traditional Sichuan brocade was composed of basic-weave, the surface of the fabric shows a regular interweaving planar texture and is difficult to represent a three-dimensional effect. Inspired by embroidery handcraft, this paper attempts to achieve the embroidery-like effect on the fabric through the jacquard process. Based on the multi-backed structure of traditional Sichuan brocade, we adopted the zoned-combination design mode and added extra free-floats interlacing weave in the area where we want to show the embroidered effect and arranged the interlacing points by referring to the feature of the pattern. As a result, designed Sichuan brocades by this method are capable of displaying embroidered effect with high realism and three-dimensionality. This approach improves the artistic effect of the traditional Sichuan brocade and provides a technical reference for further texture design of jacquard fabrics.展开更多
Primary impairments of developmental coordination disorder (DCD) include impairments in motor skill, motor learning, and imitation. Such difficulties present challenges for individuals with DCD and may persist into ad...Primary impairments of developmental coordination disorder (DCD) include impairments in motor skill, motor learning, and imitation. Such difficulties present challenges for individuals with DCD and may persist into adulthood, negatively impacting daily life in school, work, and social domains. A better understanding of the neural correlates of motor and imitation impairments in DCD holds the potential for informing development of treatment approaches to address these impairments. Although the disorder is assumed to be of neurological origin, little is known of the brain-based etiology of DCD. In recent years the discovery of a fronto-parietal circuit—known as the mirror neuron system—has enabled researchers to better understand imitation, general motor functions, and aspects of social cognition. Given its involvement in imitation and other motor functions, we propose that dysfunction in the mirror neuron system may underlie the characteristic impairments of DCD. We review literature pertaining to the mirror neuron system and develop a theory of disordered mirror neuron functioning in DCD. Finally, we review the limited neuroimaging literature available on neural correlates of DCD and show that the findings from those investigations are congruent with a mirror neuron system theory of DCD. Future research in this population should be designed to investigate specifically mirror neuron regions in individuals with DCD during skilled motor tasks and imitation in particular.展开更多
The error sources related to the laser rangefinder,GPS and INS are analyzed in details.Several coordinates systems used in airborne laser scanning are set up,and then the basic formula of system is given.This paper em...The error sources related to the laser rangefinder,GPS and INS are analyzed in details.Several coordinates systems used in airborne laser scanning are set up,and then the basic formula of system is given.This paper emphasizes on discussing the kinematic offset correction between GPS antenna phase center and laser fired point.And kinematic time delay influence on laser footprint position,the ranging errors,positioning errors,attitude errors and integration errors of the system are also explored.Finally,the result shows that the kinematic time delay can be neglected as compared with other error sources.The accuracy of the coordinates is not only influenced by the amplitude of the error,but also controlled by the operation parameters such as flight height,scanning angle amplitude and attitude magnitude of the platform.展开更多
In real life, in different industries, we often deal with systems designed for multiple use for performing single-type tasks. Processes taking place at this time are called service of requirements, and the systems the...In real life, in different industries, we often deal with systems designed for multiple use for performing single-type tasks. Processes taking place at this time are called service of requirements, and the systems themselves—Queueing Systems. This article is dedicated to computer software modelling of processes taking place in the systems in question, Markov processes in particular. In this article, by means of Matlab environment, software realization of one of the typical models of queueing service theory-multichannel QS with unreliable recoverable servers and limited number of requirements in the system, is fulfilled. The results of this research are important because it gives the possibility to use received results to determine optimality degree of some real queueing systems that possess Markov property.展开更多
Although we have no clear picture of the life of Hanshan,a legendary TANG monk and in Collected Poems of Hanshan(Hanshan Sho'i),we can find either unclear ideas regarding his major thoughts or different ideologies...Although we have no clear picture of the life of Hanshan,a legendary TANG monk and in Collected Poems of Hanshan(Hanshan Sho'i),we can find either unclear ideas regarding his major thoughts or different ideologies from Confucianism,Buddhism,and Daoism.Hanshan poetry was broadly read by people belonging to various social statuses during the SONG Dynasty.His poetry was also frequently cited in Chan Buddhist literature of the period.Furthermore,SONG Chan Buddhist monks invited Hanshan into their own genealogy and regarded him as a"San Sheng"(a Free Sage).Many Chan Buddhist monks of the SONG Dynasty used Hanshan poetry in various Chan Buddhist texts.Numerous Chan Buddhist monks even wrote so-called"ni Hanshan shi",which imitated Hanshan poetry as a kind of personal literary creation.It is understandable that when a monk imitated Hanshan poetry,he would simultaneously be both the reader and the creator of Hanshan poetry,and as we understand that every writer produces their works through their own cultural outlook,a newly-formed correlation occurred naturally between the original poetry and imitated poetry through the SONG Chan Buddhist monk's version.By observing this correlation,this paper will deeply analyze the dissemination and acceptance of Hanshan poetry,within Chan Buddhist society in the SONG Dynasty,as based on Chan Buddhist literature,in order to learn more about image creation and the recreation of Hanshan during the period.展开更多
Microsoft Kinect sensor has shown the research community that it's more than just an interactive gaming device, due to its multi-functional abilities and high reliability. In this work, online HIL (Hardware-in-the...Microsoft Kinect sensor has shown the research community that it's more than just an interactive gaming device, due to its multi-functional abilities and high reliability. In this work, online HIL (Hardware-in-the-Loop) experimental data are used to apply human motion imitation to a 2-degree of freedom Lego Mind storm NXT robotic arm. A model simulation of the dc motor used in this experiment is also present in this paper. The acquired input data from the Kinect sensor are processed in a closed loop PID controller with feedback from motors encoders. The applied algorithms solve the overlapping input problem, conducting a simultaneous control of both shoulder and elbow joints, and solving the overlapping input problem as well. The work in this paper is presented as a prototype to assure the applicability of the algorithms, for further development.展开更多
Imitation models for computing the environmental water pollution level depending on the intensity of pollution sources created by the author over the years are presented. For this purpose, an additive model of a non-s...Imitation models for computing the environmental water pollution level depending on the intensity of pollution sources created by the author over the years are presented. For this purpose, an additive model of a non-stationary random process is considered. For the modeling of its components, models that consider only dilution and self-purification processes are proposed for waste water and three-dimensional turbulent diffusion equations for river waters, and multidimensional Gaussian Markov series are proposed for modeling the random component. The purpose, the capabilities and the peculiarities of such imitation models are discussed taking into account the peculiarities of the water objects. The modular principle of creating imitation models is proposed to facilitate their development and use.展开更多
Visual servoing is an active and popular area of research among roboticists.Eventhough visual servoing techniques enhance the perfomance,the associated systems still use traditional methods for their input control.Man...Visual servoing is an active and popular area of research among roboticists.Eventhough visual servoing techniques enhance the perfomance,the associated systems still use traditional methods for their input control.Many research activities and applications have been carried out to implement effective and precise controlling of bilateral systems.This paper presents a 3D spresctroscope-based control technique for bilateral systems.The effectiveness of the available master side designs are evaluated against gesture-based techniques.Joystick control,Electromyography(EMG),Voice control,Haptic control,Exoskeleton control,Gesture and Brain Control Interface(BCI)are identified in the litreature as available bilateral inputs.In the present technnique,Leap Motion Controller(LMC)has been introduced(LMC)to extract the human hand gestures and their parameters.Then these parameters are convereted into respective joint sapce angles using the presented mathematical model.The mathematical models for fingertip mapping,inverse kinematics,dynamics and trajectory generation are implemented and studied.Wolfman Mathematica 10 and MATLAB simulation framework are used to validate the mathematical models,simulations and developed control algorithms.The developed system has sucesfully imitated the fingertip motion.In particular,the system has been able to imitate the figretip motion with a deviation of 6.7%in X axis,5.5%in Y axis and7.9%in Z axis with respect to the expected position.展开更多
The existing research on rescue robots has focused mainly on reconnaissance,detection,and firefighting,and a small number of robots that can achieve human rescue have problems such as poor safety and stability and ins...The existing research on rescue robots has focused mainly on reconnaissance,detection,and firefighting,and a small number of robots that can achieve human rescue have problems such as poor safety and stability and insufficient carrying capacity.This article addresses the above issues and cleverly combines the advantages of soft robotic arms,underactuated robotic arms,and suction cups based on the principles of bionics.A new design for a robotic arm was proposed,and its working principle was explained.Then,the human rescue process was divided into two stages,and the grasping force of the robotic arm in each stage was analyzed separately.Finally,a prototype of the principle was developed,and the feasibility of the design principle of the robotic arm was verified through grasping experiments on a cross-sectional contour model of the human chest.At the same time,grasping experiments were conducted on different objects to demonstrate the potential application of the robotic arm in grasping ground objects.This research proposes a stress envelope adsorption rescue robot arm inspired by the adhesion ability of the Drosera plant and the stress envelope effect,which can apply force to the entire surface of the human body,reduce local force on the human body,ensure load-bearing capacity and adaptability,and improve the safety and stability of rescue grasping.展开更多
Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high c...Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high computational overhead.This study proposes a lightweight integrated framework for grasp detection and imitation learning,named GD-IL;it comprises a grasp detection algorithm based on manipulability and Gaussian mixture model(manipulability-GMM),and a grasp trajectory generation algorithm based on a two-stage robot imitation learning algorithm(TS-RIL).In the manipulability-GMM algorithm,we apply GMM clustering and ellipse regression to the object point cloud,propose two judgment criteria to generate multiple candidate grasp bounding boxes for the robot,and use manipulability as a metric for selecting the optimal grasp bounding box.The stages of the TS-RIL algorithm are grasp trajectory learning and robot pose optimization.In the first stage,the robot grasp trajectory is characterized using a second-order dynamic movement primitive model and Gaussian mixture regression(GMM).By adjusting the function form of the forcing term,the robot closely approximates the target-grasping trajectory.In the second stage,a robot pose optimization model is built based on the derived pose error formula and manipulability metric.This model allows the robot to adjust its configuration in real time while grasping,thereby effectively avoiding singularities.Finally,an algorithm verification platform is developed based on a Robot Operating System and a series of comparative experiments are conducted in real-world scenarios.The experimental results demonstrate that GD-IL significantly improves the effectiveness and robustness of grasp detection and trajectory imitation learning,outperforming existing state-of-the-art methods in execution efficiency,manipulability,and success rate.展开更多
This paper explores the impact of inward foreign direct investment(FDI)on entrepreneurial activity in host countries.It argues that inward FDI affects different types of entrepreneurship in distinct ways,with its impa...This paper explores the impact of inward foreign direct investment(FDI)on entrepreneurial activity in host countries.It argues that inward FDI affects different types of entrepreneurship in distinct ways,with its impact varying depending on the level of formal institutional development in the host country.On the one hand,inward FDI’s market spillover effects increase opportunities for imitative entrepreneurship and reduce entrepreneurial risk.On the other hand,inward FDI has a mixed blessing for innovative entrepreneurship:it fuels innovation through knowledge spillovers but simultaneously intensifies competition,creating uncertainty.Using 268 observations from 59 countries between 2010 and 2018,our empirical evidence reveals a striking dichotomy:inward FDI fuels imitative entrepreneurship where formal institutions are weak,yet only fosters innovative entrepreneurship where institutions are robust.Further analysis identifies key factors-such as host country R&D investment,intellectual property protection,financial development,and entrepreneurial support-that influence whether inward FDI can effectively foster innovative entrepreneurship.展开更多
The automatic and rapid generation of excavation trajectories is the foundation for achieving an intelligent excavator.To obtain high-performance trajectories that enhance operational capacity while avoiding the numer...The automatic and rapid generation of excavation trajectories is the foundation for achieving an intelligent excavator.To obtain high-performance trajectories that enhance operational capacity while avoiding the numerous issues present in existing methods for generating effective excavation paths,this paper proposes a trajectory generation method for excavators based on imitation learning,using the mole as a bionic prototype.Given the high excavation efficiency of moles,this paper first analyzes the structural characteristics of the mole’s forelimbs,its digging principles,morphology,and trajectory patterns.Subsequently,a higher-order polynomial is employed to fit and optimize the mole’s excavation trajectory.Next,imitation learning is conducted on sample trajectories based on Dynamic Movement Primitives,followed by the introduction of an obstacle avoidance algorithm.Simulation experiments and comparisons demonstrate that the mole-inspired trajectory method used in this paper performs well and possesses the ability to generate obstacle avoidance trajectories,as well as the convenience of transferring across different machine models.展开更多
基金supported by State Grid Corporation Technology Project(No.522437250003).
文摘Hydrogen energy is a crucial support for China’s low-carbon energy transition.With the large-scale integration of renewable energy,the combination of hydrogen and integrated energy systems has become one of the most promising directions of development.This paper proposes an optimized schedulingmodel for a hydrogen-coupled electro-heat-gas integrated energy system(HCEHG-IES)using generative adversarial imitation learning(GAIL).The model aims to enhance renewable-energy absorption,reduce carbon emissions,and improve grid-regulation flexibility.First,the optimal scheduling problem of HCEHG-IES under uncertainty is modeled as a Markov decision process(MDP).To overcome the limitations of conventional deep reinforcement learning algorithms—including long optimization time,slow convergence,and subjective reward design—this study augments the PPO algorithm by incorporating a discriminator network and expert data.The newly developed algorithm,termed GAIL,enables the agent to perform imitation learning from expert data.Based on this model,dynamic scheduling decisions are made in continuous state and action spaces,generating optimal energy-allocation and management schemes.Simulation results indicate that,compared with traditional reinforcement-learning algorithms,the proposed algorithmoffers better economic performance.Guided by expert data,the agent avoids blind optimization,shortens the offline training time,and improves convergence performance.In the online phase,the algorithm enables flexible energy utilization,thereby promoting renewable-energy absorption and reducing carbon emissions.
基金the Australian Research Council (ARC) through Future Fellowship (FT210100298)Discovery Project (DP220100603)+8 种基金Linkage Project (LP210200504,LP220100088, LP230200897)Industrial Transformation Research Hub (IH240100009) schemesthe Australian Government through the Cooperative Research Centres Projects (CRCPXIII000077)the Australian Renewable Energy Agency (ARENA) as part of ARENA’s Transformative Research Accelerating Commercialisation Program (TM021)European Commission’s Australia-Spain Network for Innovation and Research Excellence (Au Spire)the Foundation of Liaoning Province Education Administration (2020LQN03)the Foundation of Liaoning Province Education Administration in 2024 (Independent topic selection-Natural science category-Strategic industrialization project LJ212410163023)the Scientific Research Fund of Liaoning Provincial Education Department (JYTMS20230767)the Liaoning Revitalization Talents Program (XLYC2007132)
文摘As a green sustainable alternative technology,synthesizing nitrate by electrocatalytic nitrogen oxidation reaction(NOR)can replace the traditional energyintensive Ostwald process.But low nitrogen fixation yields and poor selectivity due to the high bond energy of the N≡N bond and competition from the oxygen evolution reaction in the electrolyte restrict its application.On the other hand,two-dimensional(2D)PdS_(2)as a member in the family of group-10 novel transition metal dichalcogenides(NTMDs)presents the interesting optical and electronic properties due to its novel folded pentagonal structure,but few researches involve to its fabrication and application.Herein,unique imitating growth feature for PdS_(2)on different 2D substrates has been firstly discovered for constructing 2D/2D heterostructures by interface engineering.Due to the different exposed chemical groups on the substrates,PdS_(2)grows as the imitation to the morphologies of the substrates and presents different thickness,size,shape and the degree of oxidation,resulting in the significant difference in the NOR activity and stability of the obtained composite catalysts.Especially,the thin and small PdS_(2)nanoplates with more defects can be obtained by decorating poly(1-vinyl-3-ethylimidazolium bromide)on the 2D substrate,easily oxidized during the preparation process,resulting in the in situ generation of SO_(4)^(2−),which plays a crucial role in reducing the activation energy of the NOR process,leading to improved efficiency for nitrate production,verified by theoretical calculation.This research provides valuable insights for the development of novel electrocatalysts based on NTMDs for NOR and highlights the importance of interface engineering in enhancing catalytic performance.
基金supported by the Natural Science Foundation of Fujian Province of China(2025J01380)National Natural Science Foundation of China(No.62471139)+3 种基金the Major Health Research Project of Fujian Province(2021ZD01001)Fujian Provincial Units Special Funds for Education and Research(2022639)Fujian University of Technology Research Start-up Fund(GY-S24002)Fujian Research and Training Grants for Young and Middle-aged Leaders in Healthcare(GY-H-24179).
文摘The generation of synthetic trajectories has become essential in various fields for analyzing complex movement patterns.However,the use of real-world trajectory data poses significant privacy risks,such as location reidentification and correlation attacks.To address these challenges,privacy-preserving trajectory generation methods are critical for applications relying on sensitive location data.This paper introduces DPIL-Traj,an advanced framework designed to generate synthetic trajectories while achieving a superior balance between data utility and privacy preservation.Firstly,the framework incorporates Differential Privacy Clustering,which anonymizes trajectory data by applying differential privacy techniques that add noise,ensuring the protection of sensitive user information.Secondly,Imitation Learning is used to replicate decision-making behaviors observed in real-world trajectories.By learning from expert trajectories,this component generates synthetic data that closely mimics real-world decision-making processes while optimizing the quality of the generated trajectories.Finally,Markov-based Trajectory Generation is employed to capture and maintain the inherent temporal dynamics of movement patterns.Extensive experiments conducted on the GeoLife trajectory dataset show that DPIL-Traj improves utility performance by an average of 19.85%,and in terms of privacy performance by an average of 12.51%,compared to state-of-the-art approaches.Ablation studies further reveal that DP clustering effectively safeguards privacy,imitation learning enhances utility under noise,and the Markov module strengthens temporal coherence.
基金supported by the National Natural Science Foundation of China(No.62576349)。
文摘Unmanned aerial vehicles(UAVs)face the challenge of autonomous obstacle avoidance in complex,multi-obstacle environments.Behavior cloning offers a promising approach to rapidly acquire a learning policy from limited expert demonstrations.However,pure imitation learning inherently suffers from poor exploration and limited generalization,typically necessitating extensive datasets to train competent student policies.We utilize a cross-modal variational autoencoder(CM-VAE)to extract compact features from raw visual inputs and UAV states,which then feed into a policy network.We evaluated our approach in a simulated environment featuring a challenging circular trajectory with eight gate obstacles.The results demonstrate that the policy trained with pure behavior cloning consistently failed.In stark contrast,our DAgger-augmented behavior cloning method successfully traversed all gates without collision.Our findings confirm that DAgger effectively mitigates the shortcomings of behavior cloning,enabling the creation of reliable and sample-efficient navigation policies for UAVs.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.RS-2024-00406320)the Institute of Information&Communica-tions Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization Program Grant funded by the Korea government(MSIT)(IITP-2026-RS-2023-00259678).
文摘Domain adaptation aims to reduce the distribution gap between the training data(source domain)and the target data.This enables effective predictions even for domains not seen during training.However,most conventional domain adaptation methods assume a single source domain,making them less suitable for modern deep learning settings that rely on diverse and large-scale datasets.To address this limitation,recent research has focused on Multi-Source Domain Adaptation(MSDA),which aims to learn effectively from multiple source domains.In this paper,we propose Efficient Domain Transition for Multi-source(EDTM),a novel and efficient framework designed to tackle two major challenges in existing MSDA approaches:(1)integrating knowledge across different source domains and(2)aligning label distributions between source and target domains.EDTM leverages an ensemble-based classifier expert mechanism to enhance the contribution of source domains that are more similar to the target domain.To further stabilize the learning process and improve performance,we incorporate imitation learning into the training of the target model.In addition,Maximum Classifier Discrepancy(MCD)is employed to align class-wise label distributions between the source and target domains.Experiments were conducted using Digits-Five,one of the most representative benchmark datasets for MSDA.The results show that EDTM consistently outperforms existing methods in terms of average classification accuracy.Notably,EDTM achieved significantly higher performance on target domains such as Modified National Institute of Standards and Technolog with blended background images(MNIST-M)and Street View House Numbers(SVHN)datasets,demonstrating enhanced generalization compared to baseline approaches.Furthermore,an ablation study analyzing the contribution of each loss component validated the effectiveness of the framework,highlighting the importance of each module in achieving optimal performance.
文摘It is of vital importance for modern college English teaching to properly construct an interactive multimedia-internet-based teaching system, the structure of which is clearly elaborated in this paper. An IMITS usually consists of hardware, software, teaching and management. At the end of this paper, a conclusion is made that only when all the four parts of IMITS are construct ed such as is demonstrated, can the IMITS exert its full effects in college English teaching.
基金National Natural Science Foundation ofChina( No.5 9965 0 0 2)
文摘Environmental comprehensive management system, called “the bionic community”, can be established in imitation of biome, which can transform the wastes generated in a certain field into the raw materials of other field. The establishment of the bionic community includes two aspects, i.e., the matching technique and the management system. The main matching technique is the preparation of composite materials made of various wastes. This new kind of material can be divided into four types: polymer matrix, silicate matrix, metal matrix and carbon matrix(or ceramic matrix). The environmental comprehensive management system is formed by organizing a trans-trades joint-management business entity with the products of composite material made of wastes at the core.
文摘Owing to the diversity of consumer’s demand, the traditional Sichuan brocade products have to be innovated in order to survive in a competitive environment. Since the construction of traditional Sichuan brocade was composed of basic-weave, the surface of the fabric shows a regular interweaving planar texture and is difficult to represent a three-dimensional effect. Inspired by embroidery handcraft, this paper attempts to achieve the embroidery-like effect on the fabric through the jacquard process. Based on the multi-backed structure of traditional Sichuan brocade, we adopted the zoned-combination design mode and added extra free-floats interlacing weave in the area where we want to show the embroidered effect and arranged the interlacing points by referring to the feature of the pattern. As a result, designed Sichuan brocades by this method are capable of displaying embroidered effect with high realism and three-dimensionality. This approach improves the artistic effect of the traditional Sichuan brocade and provides a technical reference for further texture design of jacquard fabrics.
文摘Primary impairments of developmental coordination disorder (DCD) include impairments in motor skill, motor learning, and imitation. Such difficulties present challenges for individuals with DCD and may persist into adulthood, negatively impacting daily life in school, work, and social domains. A better understanding of the neural correlates of motor and imitation impairments in DCD holds the potential for informing development of treatment approaches to address these impairments. Although the disorder is assumed to be of neurological origin, little is known of the brain-based etiology of DCD. In recent years the discovery of a fronto-parietal circuit—known as the mirror neuron system—has enabled researchers to better understand imitation, general motor functions, and aspects of social cognition. Given its involvement in imitation and other motor functions, we propose that dysfunction in the mirror neuron system may underlie the characteristic impairments of DCD. We review literature pertaining to the mirror neuron system and develop a theory of disordered mirror neuron functioning in DCD. Finally, we review the limited neuroimaging literature available on neural correlates of DCD and show that the findings from those investigations are congruent with a mirror neuron system theory of DCD. Future research in this population should be designed to investigate specifically mirror neuron regions in individuals with DCD during skilled motor tasks and imitation in particular.
基金Funded by the National Natural Science Foundation of China(No.40004001)
文摘The error sources related to the laser rangefinder,GPS and INS are analyzed in details.Several coordinates systems used in airborne laser scanning are set up,and then the basic formula of system is given.This paper emphasizes on discussing the kinematic offset correction between GPS antenna phase center and laser fired point.And kinematic time delay influence on laser footprint position,the ranging errors,positioning errors,attitude errors and integration errors of the system are also explored.Finally,the result shows that the kinematic time delay can be neglected as compared with other error sources.The accuracy of the coordinates is not only influenced by the amplitude of the error,but also controlled by the operation parameters such as flight height,scanning angle amplitude and attitude magnitude of the platform.
文摘In real life, in different industries, we often deal with systems designed for multiple use for performing single-type tasks. Processes taking place at this time are called service of requirements, and the systems themselves—Queueing Systems. This article is dedicated to computer software modelling of processes taking place in the systems in question, Markov processes in particular. In this article, by means of Matlab environment, software realization of one of the typical models of queueing service theory-multichannel QS with unreliable recoverable servers and limited number of requirements in the system, is fulfilled. The results of this research are important because it gives the possibility to use received results to determine optimality degree of some real queueing systems that possess Markov property.
文摘Although we have no clear picture of the life of Hanshan,a legendary TANG monk and in Collected Poems of Hanshan(Hanshan Sho'i),we can find either unclear ideas regarding his major thoughts or different ideologies from Confucianism,Buddhism,and Daoism.Hanshan poetry was broadly read by people belonging to various social statuses during the SONG Dynasty.His poetry was also frequently cited in Chan Buddhist literature of the period.Furthermore,SONG Chan Buddhist monks invited Hanshan into their own genealogy and regarded him as a"San Sheng"(a Free Sage).Many Chan Buddhist monks of the SONG Dynasty used Hanshan poetry in various Chan Buddhist texts.Numerous Chan Buddhist monks even wrote so-called"ni Hanshan shi",which imitated Hanshan poetry as a kind of personal literary creation.It is understandable that when a monk imitated Hanshan poetry,he would simultaneously be both the reader and the creator of Hanshan poetry,and as we understand that every writer produces their works through their own cultural outlook,a newly-formed correlation occurred naturally between the original poetry and imitated poetry through the SONG Chan Buddhist monk's version.By observing this correlation,this paper will deeply analyze the dissemination and acceptance of Hanshan poetry,within Chan Buddhist society in the SONG Dynasty,as based on Chan Buddhist literature,in order to learn more about image creation and the recreation of Hanshan during the period.
文摘Microsoft Kinect sensor has shown the research community that it's more than just an interactive gaming device, due to its multi-functional abilities and high reliability. In this work, online HIL (Hardware-in-the-Loop) experimental data are used to apply human motion imitation to a 2-degree of freedom Lego Mind storm NXT robotic arm. A model simulation of the dc motor used in this experiment is also present in this paper. The acquired input data from the Kinect sensor are processed in a closed loop PID controller with feedback from motors encoders. The applied algorithms solve the overlapping input problem, conducting a simultaneous control of both shoulder and elbow joints, and solving the overlapping input problem as well. The work in this paper is presented as a prototype to assure the applicability of the algorithms, for further development.
文摘Imitation models for computing the environmental water pollution level depending on the intensity of pollution sources created by the author over the years are presented. For this purpose, an additive model of a non-stationary random process is considered. For the modeling of its components, models that consider only dilution and self-purification processes are proposed for waste water and three-dimensional turbulent diffusion equations for river waters, and multidimensional Gaussian Markov series are proposed for modeling the random component. The purpose, the capabilities and the peculiarities of such imitation models are discussed taking into account the peculiarities of the water objects. The modular principle of creating imitation models is proposed to facilitate their development and use.
文摘Visual servoing is an active and popular area of research among roboticists.Eventhough visual servoing techniques enhance the perfomance,the associated systems still use traditional methods for their input control.Many research activities and applications have been carried out to implement effective and precise controlling of bilateral systems.This paper presents a 3D spresctroscope-based control technique for bilateral systems.The effectiveness of the available master side designs are evaluated against gesture-based techniques.Joystick control,Electromyography(EMG),Voice control,Haptic control,Exoskeleton control,Gesture and Brain Control Interface(BCI)are identified in the litreature as available bilateral inputs.In the present technnique,Leap Motion Controller(LMC)has been introduced(LMC)to extract the human hand gestures and their parameters.Then these parameters are convereted into respective joint sapce angles using the presented mathematical model.The mathematical models for fingertip mapping,inverse kinematics,dynamics and trajectory generation are implemented and studied.Wolfman Mathematica 10 and MATLAB simulation framework are used to validate the mathematical models,simulations and developed control algorithms.The developed system has sucesfully imitated the fingertip motion.In particular,the system has been able to imitate the figretip motion with a deviation of 6.7%in X axis,5.5%in Y axis and7.9%in Z axis with respect to the expected position.
基金Supported by National Natural Science Foundation of China(Grant No.52475032)Central Government Guides Local Science and Technology Development Fund Projects(Grant No.246Z2001G)Hebei Provincial Natural Science Foundation Key Projects(Grant No.E2021203125).
文摘The existing research on rescue robots has focused mainly on reconnaissance,detection,and firefighting,and a small number of robots that can achieve human rescue have problems such as poor safety and stability and insufficient carrying capacity.This article addresses the above issues and cleverly combines the advantages of soft robotic arms,underactuated robotic arms,and suction cups based on the principles of bionics.A new design for a robotic arm was proposed,and its working principle was explained.Then,the human rescue process was divided into two stages,and the grasping force of the robotic arm in each stage was analyzed separately.Finally,a prototype of the principle was developed,and the feasibility of the design principle of the robotic arm was verified through grasping experiments on a cross-sectional contour model of the human chest.At the same time,grasping experiments were conducted on different objects to demonstrate the potential application of the robotic arm in grasping ground objects.This research proposes a stress envelope adsorption rescue robot arm inspired by the adhesion ability of the Drosera plant and the stress envelope effect,which can apply force to the entire surface of the human body,reduce local force on the human body,ensure load-bearing capacity and adaptability,and improve the safety and stability of rescue grasping.
基金Supported by National Natural Science Foundation of China(Grant No.52475280)Shaanxi Provincial Natural Science Basic Research Program(Grant No.2025SYSSYSZD-105).
文摘Robots are key to expanding the scope of space applications.The end-to-end training for robot vision-based detection and precision operations is challenging owing to constraints such as extreme environments and high computational overhead.This study proposes a lightweight integrated framework for grasp detection and imitation learning,named GD-IL;it comprises a grasp detection algorithm based on manipulability and Gaussian mixture model(manipulability-GMM),and a grasp trajectory generation algorithm based on a two-stage robot imitation learning algorithm(TS-RIL).In the manipulability-GMM algorithm,we apply GMM clustering and ellipse regression to the object point cloud,propose two judgment criteria to generate multiple candidate grasp bounding boxes for the robot,and use manipulability as a metric for selecting the optimal grasp bounding box.The stages of the TS-RIL algorithm are grasp trajectory learning and robot pose optimization.In the first stage,the robot grasp trajectory is characterized using a second-order dynamic movement primitive model and Gaussian mixture regression(GMM).By adjusting the function form of the forcing term,the robot closely approximates the target-grasping trajectory.In the second stage,a robot pose optimization model is built based on the derived pose error formula and manipulability metric.This model allows the robot to adjust its configuration in real time while grasping,thereby effectively avoiding singularities.Finally,an algorithm verification platform is developed based on a Robot Operating System and a series of comparative experiments are conducted in real-world scenarios.The experimental results demonstrate that GD-IL significantly improves the effectiveness and robustness of grasp detection and trajectory imitation learning,outperforming existing state-of-the-art methods in execution efficiency,manipulability,and success rate.
基金supported by the National Natural Science Foundation of China(NSFC)“Research on Entrepreneurial Activities in Dynamic and Complex Institutional Environments”(Grant No.71872193)the Guangdong Basic and Applied Basic Research Foundation“Research on the Multiple Institutional Effects of Innovative Entrepreneurship from the Perspective of Knowledge Utilization”(Grant No.2023A1515110323)the Guangdong Basic and Applied Basic Research Foundation“From Strategic to Substantive Transformation:The Dynamic Process of Multi-Factor Family Involvement and Enterprise Digital Transformation”(Grant No.2024A1515012619).
文摘This paper explores the impact of inward foreign direct investment(FDI)on entrepreneurial activity in host countries.It argues that inward FDI affects different types of entrepreneurship in distinct ways,with its impact varying depending on the level of formal institutional development in the host country.On the one hand,inward FDI’s market spillover effects increase opportunities for imitative entrepreneurship and reduce entrepreneurial risk.On the other hand,inward FDI has a mixed blessing for innovative entrepreneurship:it fuels innovation through knowledge spillovers but simultaneously intensifies competition,creating uncertainty.Using 268 observations from 59 countries between 2010 and 2018,our empirical evidence reveals a striking dichotomy:inward FDI fuels imitative entrepreneurship where formal institutions are weak,yet only fosters innovative entrepreneurship where institutions are robust.Further analysis identifies key factors-such as host country R&D investment,intellectual property protection,financial development,and entrepreneurial support-that influence whether inward FDI can effectively foster innovative entrepreneurship.
基金supported by the National Science Foundation of China(Grant No.52375246,No.52372428,No.52105100)Guangxi Science and Technology Program(Grant No.2023AB09014)Jilin Province Science and Technology Development Program,(Grant No.20230201094GX,No.20230201069GX).
文摘The automatic and rapid generation of excavation trajectories is the foundation for achieving an intelligent excavator.To obtain high-performance trajectories that enhance operational capacity while avoiding the numerous issues present in existing methods for generating effective excavation paths,this paper proposes a trajectory generation method for excavators based on imitation learning,using the mole as a bionic prototype.Given the high excavation efficiency of moles,this paper first analyzes the structural characteristics of the mole’s forelimbs,its digging principles,morphology,and trajectory patterns.Subsequently,a higher-order polynomial is employed to fit and optimize the mole’s excavation trajectory.Next,imitation learning is conducted on sample trajectories based on Dynamic Movement Primitives,followed by the introduction of an obstacle avoidance algorithm.Simulation experiments and comparisons demonstrate that the mole-inspired trajectory method used in this paper performs well and possesses the ability to generate obstacle avoidance trajectories,as well as the convenience of transferring across different machine models.
