The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fau...The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fault characteristics under different loads is markedly inconsistent,and data is hard to label,which makes it difficult for the traditional diagnosis method based on single-condition training to generalize to different conditions.To address these issues,the paper proposes a novel transfer discriminant neural network(TDNN)for gear fault diagnosis.Specifically,an optimized joint distribution adaptive mechanism(OJDA)is designed to solve the distribution alignment problem between two domains.To improve the classification effect within the domain and the feature recognition capability for a few labeled data,metric learning is introduced to distinguish features from different fault categories.In addition,TDNN adopts a new pseudo-label training strategy to achieve label replacement by comparing the maximum probability of the pseudo-label with the test result.The proposed TDNN is verified in the experimental data set of the artillery manipulator device,and the diagnosis can achieve 99.5%,significantly outperforming other traditional adaptation methods.展开更多
Solar radio burst(SRB)is one of the main natural interference sources of Global Positioning System(GPS)signals and can reduce the signal-to-noise ratio(SNR),directly affecting the tracking performance of GPS receivers...Solar radio burst(SRB)is one of the main natural interference sources of Global Positioning System(GPS)signals and can reduce the signal-to-noise ratio(SNR),directly affecting the tracking performance of GPS receivers.In this paper,a tracking algorithm based on the adaptive Kalman filter(AKF)with carrier-to-noise ratio estimation is proposed and compared with the conventional second-order phase-locked loop tracking algo-rithms and the improved Sage-Husa adaptive Kalman filter(SHAKF)algorithm.It is discovered that when the SRBs occur,the improved SHAKF and the AKF with carrier-to-noise ratio estimation enable stable tracking to loop signals.The conven-tional second-order phase-locked loop tracking algorithms fail to track the receiver signal.The standard deviation of the carrier phase error of the AKF with carrier-to-noise ratio estimation out-performs 50.51%of the improved SHAKF algorithm,showing less fluctuation and better stability.The proposed algorithm is proven to show more excellent adaptability in the severe envi-ronment caused by the SRB occurrence and has better tracking performance.展开更多
Since backlash nonlinearity is inevitably existing in actuators for bidirectional stabilization system of allelectric tank,it behaves more drastically in high maneuvering environments.In this work,the accurate trackin...Since backlash nonlinearity is inevitably existing in actuators for bidirectional stabilization system of allelectric tank,it behaves more drastically in high maneuvering environments.In this work,the accurate tracking control for bidirectional stabilization system of moving all-electric tank with actuator backlash and unmodeled disturbance is solved.By utilizing the smooth adaptive backlash inverse model,a nonlinear robust adaptive feedback control scheme is presented.The unknown parameters and unmodelled disturbance are addressed separately through the derived parametric adaptive function and the continuous nonlinear robust term.Because the unknown backlash parameters are updated via adaptive function and the backlash effect can be suppressed successfully by inverse operation,which ensures the system stability.Meanwhile,the system disturbance in the high maneuverable environment can be estimated with the constructed adaptive law online improving the engineering practicality.Finally,Lyapunov-based analysis proves that the developed controller can ensure the tracking error asymptotically converges to zero even with unmodeled disturbance and unknown actuator backlash.Contrast co-simulations and experiments illustrate the advantages of the proposed approach.展开更多
This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy ...This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.展开更多
This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a larg...This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a large class of engineering systems,such as vehicular systems,robot manipulators and satellites.All these systems are often characterized by highly nonlinear characteristics,heavy modeling uncertainties and unknown perturbations,therefore,accurate-model-based nonlinear control approaches become unavailable.Motivated by the challenge,a reinforcement learning(RL)adaptive control methodology based on the actor-critic framework is investigated to compensate the uncertain mechanical dynamics.The approximation inaccuracies caused by RL and the exogenous unknown disturbances are circumvented via a continuous robust integral of the sign of the error(RISE)control approach.Different from a classical RISE control law,a tanh(·)function is utilized instead of a sign(·)function to acquire a more smooth control signal.The developed controller requires very little prior knowledge of the dynamic model,is robust to unknown dynamics and exogenous disturbances,and can achieve asymptotic output tracking.Eventually,co-simulations through ADAMS and MATLAB/Simulink on a three degrees-of-freedom(3-DOF)manipulator and experiments on a real-time electromechanical servo system are performed to verify the performance of the proposed approach.展开更多
As for ultra-low permeability reservoir,the adaptability of common nine-spot well pattern is studied through large-scale flat models made by micro-fractured natural sandstone outcrops.Combined with non-linear porous f...As for ultra-low permeability reservoir,the adaptability of common nine-spot well pattern is studied through large-scale flat models made by micro-fractured natural sandstone outcrops.Combined with non-linear porous flow characteristics,the concept of dimensionless pressure sweep efficiency and deliverability index are put forward to evaluate the physical models' well pattern adaptability.Through experiments,the models' pressure distribution is measured and on which basis,the pressure gradient fields are drawn and the porous flow regions of these models are divided into dead oil region,non-linear porous flow region,and quasi-linear porous flow region with the help of twin-core non-linear porous flow curve.The results indicate that rectangular well pattern in fracture reservoirs has the best adaptability,while the worst is inverted nine-spot equilateral well pattern.With the increase of drawdown pressure,dead oil region decreases,pressure sweep efficiency and deliverability index increase; meantime,the deliverability index of rectangular well pattern has much more rational increase.Under the same drawdown pressure,the rectangular well pattern has the largest pressure sweep efficiency.展开更多
In engineering application,there is only one adaptive weights estimated by most of traditional early warning radars for adaptive interference suppression in a pulse reputation interval(PRI).Therefore,if the training s...In engineering application,there is only one adaptive weights estimated by most of traditional early warning radars for adaptive interference suppression in a pulse reputation interval(PRI).Therefore,if the training samples used to calculate the weight vector does not contain the jamming,then the jamming cannot be removed by adaptive spatial filtering.If the weight vector is constantly updated in the range dimension,the training data may contain target echo signals,resulting in signal cancellation effect.To cope with the situation that the training samples are contaminated by target signal,an iterative training sample selection method based on non-homogeneous detector(NHD)is proposed in this paper for updating the weight vector in entire range dimension.The principle is presented,and the validity is proven by simulation results.展开更多
The identification of intercepted radio fuze modulation types is a prerequisite for decision-making in interference systems.However,the electromagnetic environment of modern battlefields is complex,and the signal-to-n...The identification of intercepted radio fuze modulation types is a prerequisite for decision-making in interference systems.However,the electromagnetic environment of modern battlefields is complex,and the signal-to-noise ratio(SNR)of such environments is usually low,which makes it difficult to implement accurate recognition of radio fuzes.To solve the above problem,a radio fuze automatic modulation recognition(AMR)method for low-SNR environments is proposed.First,an adaptive denoising algorithm based on data rearrangement and the two-dimensional(2D)fast Fourier transform(FFT)(DR2D)is used to reduce the noise of the intercepted radio fuze intermediate frequency(IF)signal.Then,the textural features of the denoised IF signal rearranged data matrix are extracted from the statistical indicator vectors of gray-level cooccurrence matrices(GLCMs),and support vector machines(SVMs)are used for classification.The DR2D-based adaptive denoising algorithm achieves an average correlation coefficient of more than 0.76 for ten fuze types under SNRs of-10 d B and above,which is higher than that of other typical algorithms.The trained SVM classification model achieves an average recognition accuracy of more than 96%on seven modulation types and recognition accuracies of more than 94%on each modulation type under SNRs of-12 d B and above,which represents a good AMR performance of radio fuzes under low SNRs.展开更多
This paper investigates the adaptive trajectory tracking control problem and the unknown parameter identification problem of a class of rotor-missiles with parametric system uncertainties.First,considering the uncerta...This paper investigates the adaptive trajectory tracking control problem and the unknown parameter identification problem of a class of rotor-missiles with parametric system uncertainties.First,considering the uncertainty of structural and aerodynamic parameters,the six-degree-of-freedom(6Do F) nonlinear equations describing the position and attitude dynamics of the rotor-missile are established,respectively,in the inertial and body-fixed reference frames.Next,a hierarchical adaptive trajectory tracking controller that can guarantee closed-loop stability is proposed according to the cascade characteristics of the 6Do F dynamics.Then,a memory-augmented update rule of unknown parameters is proposed by integrating all historical data of the regression matrix.As long as the finitely excited condition is satisfied,the precise identification of unknown parameters can be achieved.Finally,the validity of the proposed trajectory tracking controller and the parameter identification method is proved through Lyapunov stability theory and numerical simulations.展开更多
We consider a single server constant retrial queue,in which a state-dependent service policy is used to control the service rate.Customer arrival follows Poisson process,while service time and retrial time are exponen...We consider a single server constant retrial queue,in which a state-dependent service policy is used to control the service rate.Customer arrival follows Poisson process,while service time and retrial time are exponential distributions.Whenever the server is available,it admits the retrial customers into service based on a first-come first-served rule.The service rate adjusts in real-time based on the retrial queue length.An iterative algorithm is proposed to numerically solve the personal optimal problem in the fully observable scenario.Furthermore,we investigate the impact of parameters on the social optimal threshold.The effectiveness of the results is illustrated by two examples.展开更多
四川大学计算机学院学生团队在大规模语言模型参数高效微调系统研究方向取得重要进展,其研究成果“mLoRA:Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs”在国际数据库学术会议VLDB 2025 Rese...四川大学计算机学院学生团队在大规模语言模型参数高效微调系统研究方向取得重要进展,其研究成果“mLoRA:Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs”在国际数据库学术会议VLDB 2025 Research Track正式发表。VLDB(International Conference on Very Large Data Bases)是数据库领域的重要国际学术会议之一,涵盖数据库管理系统、数据密集型系统与大规模数据处理等方向。该工作已在多个国内外互联网企业的实际生产环境中部署应用,并获得一项中国发明专利和一项美国发明专利的受理。展开更多
Shock wave caused by a sudden release of high-energy,such as explosion and blast,usually affects a significant range of areas.The utilization of a uniform fine mesh to capture sharp shock wave and to obtain precise re...Shock wave caused by a sudden release of high-energy,such as explosion and blast,usually affects a significant range of areas.The utilization of a uniform fine mesh to capture sharp shock wave and to obtain precise results is inefficient in terms of computational resource.This is particularly evident when large-scale fluid field simulations are conducted with significant differences in computational domain size.In this work,a variable-domain-size adaptive mesh enlargement(vAME)method is developed based on the proposed adaptive mesh enlargement(AME)method for modeling multi-explosives explosion problems.The vAME method reduces the division of numerous empty areas or unnecessary computational domains by adaptively suspending enlargement operation in one or two directions,rather than in all directions as in AME method.A series of numerical tests via AME and vAME with varying nonintegral enlargement ratios and different mesh numbers are simulated to verify the efficiency and order of accuracy.An estimate of speedup ratio is analyzed for further efficiency comparison.Several large-scale near-ground explosion experiments with single/multiple explosives are performed to analyze the shock wave superposition formed by the incident wave,reflected wave,and Mach wave.Additionally,the vAME method is employed to validate the accuracy,as well as to investigate the performance of the fluid field and shock wave propagation,considering explosive quantities ranging from 1 to 5 while maintaining a constant total mass.The results show a satisfactory correlation between the overpressure versus time curves for experiments and numerical simulations.The vAME method yields a competitive efficiency,increasing the computational speed to 3.0 and approximately 120,000 times in comparison to AME and the fully fine mesh method,respectively.It indicates that the vAME method reduces the computational cost with minimal impact on the results for such large-scale high-energy release problems with significant differences in computational domain size.展开更多
Enhancing the stability and performance of practical control systems in the presence of nonlinearity,time delay,and uncertainty remains a significant challenge.Particularly,a class of strict-feedback nonlinear uncerta...Enhancing the stability and performance of practical control systems in the presence of nonlinearity,time delay,and uncertainty remains a significant challenge.Particularly,a class of strict-feedback nonlinear uncertain systems characterized by unknown control directions and time-varying input delay lacks comprehensive solutions.In this paper,we propose an observerbased adaptive tracking controller to address this gap.Neural networks are utilized to handle uncertainty,and a unique coordinate transformation is employed to untangle the coupling between input delay and unknown control directions.Subsequently,a new auxiliary signal counters the impact of time-varying input delay,while a Nussbaum function is introduced to solve the problem of unknown control directions.The leverage of an advanced dynamic surface control technique avoids the“complexity explosion”and reduces boundary layer errors.Synthesizing these techniques ensures that all the closed-loop signals are semi-globally uniformly ultimately bounded(SGUUB),and the tracking error converges to a small region around the origin by selecting suitable parameters.Simulation examples are provided to demonstrate the feasibility of the proposed approach.展开更多
In complex environments such as high dynamics and weak signals,a satellite signal compensation method based on prefabricated trajectory assistance and an improved adaptive Kalman filter is proposed for a 155 mm differ...In complex environments such as high dynamics and weak signals,a satellite signal compensation method based on prefabricated trajectory assistance and an improved adaptive Kalman filter is proposed for a 155 mm differential rotating rear-body control-guided projectile to address the situation of satellite signal flickering and loss in projectile navigation systems due to environmental limitations.First,establish the system state and measurement equation when receiving satellite signals normally.Second,a seven-degree-of-freedom external ballistic model is constructed,and the ideal trajectory output from the ballistic model is used to provide the virtual motion state of the projectile,which is input into a filter as a substitute observation when satellite signals are lost.Finally,an adaptive Kalman filter(AKF)is designed,the proposed adaptive Kalman filter can accurately adjust the estimation error covariance matrix and Kalman gain in real-time based on information covariance mismatch.The simulation results show that compared to the classical Kalman filter,it can reduce the average positioning error by more than 38.21%in the case of short-term and full-range loss of satellite signals,providing a new idea for the integrated navigation of projectiles with incomplete information under the condition of satellite signal loss.展开更多
Range-azimuth imaging of ground targets via frequency-modulated continuous wave(FMCW)radar is crucial for effective target detection.However,when the pitch of the moving array constructed during motion exceeds the phy...Range-azimuth imaging of ground targets via frequency-modulated continuous wave(FMCW)radar is crucial for effective target detection.However,when the pitch of the moving array constructed during motion exceeds the physical array aperture,azimuth ambiguity occurs,making range-azimuth imaging on a moving platform challenging.To address this issue,we theoretically analyze azimuth ambiguity generation in sparse motion arrays and propose a dual-aperture adaptive processing(DAAP)method for suppressing azimuth ambiguity.This method combines spatial multiple-input multiple-output(MIMO)arrays with sparse motion arrays to achieve high-resolution range-azimuth imaging.In addition,an adaptive QR decomposition denoising method for sparse array signals based on iterative low-rank matrix approximation(LRMA)and regularized QR is proposed to preprocess sparse motion array signals.Simulations and experiments show that on a two-transmitter-four-receiver array,the signal-to-noise ratio(SNR)of the sparse motion array signal after noise suppression via adaptive QR decomposition can exceed 0 dB,and the azimuth ambiguity signal ratio(AASR)can be reduced to below-20 dB.展开更多
With the rapid development of low-altitude economy and unmanned aerial vehicles (UAVs) deployment technology, aerial-ground collaborative delivery (AGCD) is emerging as a novel mode of last-mile delivery, where the ve...With the rapid development of low-altitude economy and unmanned aerial vehicles (UAVs) deployment technology, aerial-ground collaborative delivery (AGCD) is emerging as a novel mode of last-mile delivery, where the vehicle and its onboard UAVs are utilized efficiently. Vehicles not only provide delivery services to customers but also function as mobile ware-houses and launch/recovery platforms for UAVs. This paper addresses the vehicle routing problem with UAVs considering time window and UAV multi-delivery (VRPU-TW&MD). A mixed integer linear programming (MILP) model is developed to mini-mize delivery costs while incorporating constraints related to UAV energy consumption. Subsequently, a micro-evolution aug-mented large neighborhood search (MEALNS) algorithm incor-porating adaptive large neighborhood search (ALNS) and micro-evolution mechanism is proposed. Numerical experiments demonstrate the effectiveness of both the model and algorithm in solving the VRPU-TW&MD. The impact of key parameters on delivery performance is explored by sensitivity analysis.展开更多
As commercial drone delivery becomes increasingly popular,the extension of the vehicle routing problem with drones(VRPD)is emerging as an optimization problem of inter-ests.This paper studies a variant of VRPD in mult...As commercial drone delivery becomes increasingly popular,the extension of the vehicle routing problem with drones(VRPD)is emerging as an optimization problem of inter-ests.This paper studies a variant of VRPD in multi-trip and multi-drop(VRP-mmD).The problem aims at making schedules for the trucks and drones such that the total travel time is minimized.This paper formulate the problem with a mixed integer program-ming model and propose a two-phase algorithm,i.e.,a parallel route construction heuristic(PRCH)for the first phase and an adaptive neighbor searching heuristic(ANSH)for the second phase.The PRCH generates an initial solution by con-currently assigning as many nodes as possible to the truck–drone pair to progressively reduce the waiting time at the rendezvous node in the first phase.Then the ANSH improves the initial solution by adaptively exploring the neighborhoods in the second phase.Numerical tests on some benchmark data are conducted to verify the performance of the algorithm.The results show that the proposed algorithm can found better solu-tions than some state-of-the-art methods for all instances.More-over,an extensive analysis highlights the stability of the pro-posed algorithm.展开更多
文摘The ammunition loading system manipulator is susceptible to gear failure due to high-frequency,heavyload reciprocating motions and the absence of protective gear components.After a fault occurs,the distribution of fault characteristics under different loads is markedly inconsistent,and data is hard to label,which makes it difficult for the traditional diagnosis method based on single-condition training to generalize to different conditions.To address these issues,the paper proposes a novel transfer discriminant neural network(TDNN)for gear fault diagnosis.Specifically,an optimized joint distribution adaptive mechanism(OJDA)is designed to solve the distribution alignment problem between two domains.To improve the classification effect within the domain and the feature recognition capability for a few labeled data,metric learning is introduced to distinguish features from different fault categories.In addition,TDNN adopts a new pseudo-label training strategy to achieve label replacement by comparing the maximum probability of the pseudo-label with the test result.The proposed TDNN is verified in the experimental data set of the artillery manipulator device,and the diagnosis can achieve 99.5%,significantly outperforming other traditional adaptation methods.
基金supported by the Foundation of Key Laboratory of Micro-inertial Instrument and Advanced Navigation Technology,Ministry of Education,Chinathe National Natural Science Foundation of China (61873064)
文摘Solar radio burst(SRB)is one of the main natural interference sources of Global Positioning System(GPS)signals and can reduce the signal-to-noise ratio(SNR),directly affecting the tracking performance of GPS receivers.In this paper,a tracking algorithm based on the adaptive Kalman filter(AKF)with carrier-to-noise ratio estimation is proposed and compared with the conventional second-order phase-locked loop tracking algo-rithms and the improved Sage-Husa adaptive Kalman filter(SHAKF)algorithm.It is discovered that when the SRBs occur,the improved SHAKF and the AKF with carrier-to-noise ratio estimation enable stable tracking to loop signals.The conven-tional second-order phase-locked loop tracking algorithms fail to track the receiver signal.The standard deviation of the carrier phase error of the AKF with carrier-to-noise ratio estimation out-performs 50.51%of the improved SHAKF algorithm,showing less fluctuation and better stability.The proposed algorithm is proven to show more excellent adaptability in the severe envi-ronment caused by the SRB occurrence and has better tracking performance.
基金the National Natural Science Foundation of China(No.52275062)and(No.52075262).
文摘Since backlash nonlinearity is inevitably existing in actuators for bidirectional stabilization system of allelectric tank,it behaves more drastically in high maneuvering environments.In this work,the accurate tracking control for bidirectional stabilization system of moving all-electric tank with actuator backlash and unmodeled disturbance is solved.By utilizing the smooth adaptive backlash inverse model,a nonlinear robust adaptive feedback control scheme is presented.The unknown parameters and unmodelled disturbance are addressed separately through the derived parametric adaptive function and the continuous nonlinear robust term.Because the unknown backlash parameters are updated via adaptive function and the backlash effect can be suppressed successfully by inverse operation,which ensures the system stability.Meanwhile,the system disturbance in the high maneuverable environment can be estimated with the constructed adaptive law online improving the engineering practicality.Finally,Lyapunov-based analysis proves that the developed controller can ensure the tracking error asymptotically converges to zero even with unmodeled disturbance and unknown actuator backlash.Contrast co-simulations and experiments illustrate the advantages of the proposed approach.
基金The National Natural Science Foundation of China (32371993)The Natural Science Research Key Project of Anhui Provincial University(2022AH040125&2023AH040135)The Key Research and Development Plan of Anhui Province (202204c06020022&2023n06020057)。
文摘This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.
基金supported in part by the National Key R&D Program of China under Grant 2021YFB2011300the National Natural Science Foundation of China under Grant 52075262。
文摘This paper mainly focuses on the development of a learning-based controller for a class of uncertain mechanical systems modeled by the Euler-Lagrange formulation.The considered system can depict the behavior of a large class of engineering systems,such as vehicular systems,robot manipulators and satellites.All these systems are often characterized by highly nonlinear characteristics,heavy modeling uncertainties and unknown perturbations,therefore,accurate-model-based nonlinear control approaches become unavailable.Motivated by the challenge,a reinforcement learning(RL)adaptive control methodology based on the actor-critic framework is investigated to compensate the uncertain mechanical dynamics.The approximation inaccuracies caused by RL and the exogenous unknown disturbances are circumvented via a continuous robust integral of the sign of the error(RISE)control approach.Different from a classical RISE control law,a tanh(·)function is utilized instead of a sign(·)function to acquire a more smooth control signal.The developed controller requires very little prior knowledge of the dynamic model,is robust to unknown dynamics and exogenous disturbances,and can achieve asymptotic output tracking.Eventually,co-simulations through ADAMS and MATLAB/Simulink on a three degrees-of-freedom(3-DOF)manipulator and experiments on a real-time electromechanical servo system are performed to verify the performance of the proposed approach.
基金Project(2011ZX05013-006)supported by the National Science and Technology Project of China
文摘As for ultra-low permeability reservoir,the adaptability of common nine-spot well pattern is studied through large-scale flat models made by micro-fractured natural sandstone outcrops.Combined with non-linear porous flow characteristics,the concept of dimensionless pressure sweep efficiency and deliverability index are put forward to evaluate the physical models' well pattern adaptability.Through experiments,the models' pressure distribution is measured and on which basis,the pressure gradient fields are drawn and the porous flow regions of these models are divided into dead oil region,non-linear porous flow region,and quasi-linear porous flow region with the help of twin-core non-linear porous flow curve.The results indicate that rectangular well pattern in fracture reservoirs has the best adaptability,while the worst is inverted nine-spot equilateral well pattern.With the increase of drawdown pressure,dead oil region decreases,pressure sweep efficiency and deliverability index increase; meantime,the deliverability index of rectangular well pattern has much more rational increase.Under the same drawdown pressure,the rectangular well pattern has the largest pressure sweep efficiency.
基金supported by the National Natural Science Foundation of China(62371049)。
文摘In engineering application,there is only one adaptive weights estimated by most of traditional early warning radars for adaptive interference suppression in a pulse reputation interval(PRI).Therefore,if the training samples used to calculate the weight vector does not contain the jamming,then the jamming cannot be removed by adaptive spatial filtering.If the weight vector is constantly updated in the range dimension,the training data may contain target echo signals,resulting in signal cancellation effect.To cope with the situation that the training samples are contaminated by target signal,an iterative training sample selection method based on non-homogeneous detector(NHD)is proposed in this paper for updating the weight vector in entire range dimension.The principle is presented,and the validity is proven by simulation results.
基金National Natural Science Foundation of China under Grant No.61973037China Postdoctoral Science Foundation 2022M720419 to provide fund for conducting experiments。
文摘The identification of intercepted radio fuze modulation types is a prerequisite for decision-making in interference systems.However,the electromagnetic environment of modern battlefields is complex,and the signal-to-noise ratio(SNR)of such environments is usually low,which makes it difficult to implement accurate recognition of radio fuzes.To solve the above problem,a radio fuze automatic modulation recognition(AMR)method for low-SNR environments is proposed.First,an adaptive denoising algorithm based on data rearrangement and the two-dimensional(2D)fast Fourier transform(FFT)(DR2D)is used to reduce the noise of the intercepted radio fuze intermediate frequency(IF)signal.Then,the textural features of the denoised IF signal rearranged data matrix are extracted from the statistical indicator vectors of gray-level cooccurrence matrices(GLCMs),and support vector machines(SVMs)are used for classification.The DR2D-based adaptive denoising algorithm achieves an average correlation coefficient of more than 0.76 for ten fuze types under SNRs of-10 d B and above,which is higher than that of other typical algorithms.The trained SVM classification model achieves an average recognition accuracy of more than 96%on seven modulation types and recognition accuracies of more than 94%on each modulation type under SNRs of-12 d B and above,which represents a good AMR performance of radio fuzes under low SNRs.
基金partially supported by the Natural Science Foundation of China (Grant Nos.62103052,52272358)partially supported by the Beijing Institute of Technology Research Fund Program for Young Scholars。
文摘This paper investigates the adaptive trajectory tracking control problem and the unknown parameter identification problem of a class of rotor-missiles with parametric system uncertainties.First,considering the uncertainty of structural and aerodynamic parameters,the six-degree-of-freedom(6Do F) nonlinear equations describing the position and attitude dynamics of the rotor-missile are established,respectively,in the inertial and body-fixed reference frames.Next,a hierarchical adaptive trajectory tracking controller that can guarantee closed-loop stability is proposed according to the cascade characteristics of the 6Do F dynamics.Then,a memory-augmented update rule of unknown parameters is proposed by integrating all historical data of the regression matrix.As long as the finitely excited condition is satisfied,the precise identification of unknown parameters can be achieved.Finally,the validity of the proposed trajectory tracking controller and the parameter identification method is proved through Lyapunov stability theory and numerical simulations.
基金supported by the National Natural Science Foundation of China(Grant No.11971486)。
文摘We consider a single server constant retrial queue,in which a state-dependent service policy is used to control the service rate.Customer arrival follows Poisson process,while service time and retrial time are exponential distributions.Whenever the server is available,it admits the retrial customers into service based on a first-come first-served rule.The service rate adjusts in real-time based on the retrial queue length.An iterative algorithm is proposed to numerically solve the personal optimal problem in the fully observable scenario.Furthermore,we investigate the impact of parameters on the social optimal threshold.The effectiveness of the results is illustrated by two examples.
文摘四川大学计算机学院学生团队在大规模语言模型参数高效微调系统研究方向取得重要进展,其研究成果“mLoRA:Fine-Tuning LoRA Adapters via Highly-Efficient Pipeline Parallelism in Multiple GPUs”在国际数据库学术会议VLDB 2025 Research Track正式发表。VLDB(International Conference on Very Large Data Bases)是数据库领域的重要国际学术会议之一,涵盖数据库管理系统、数据密集型系统与大规模数据处理等方向。该工作已在多个国内外互联网企业的实际生产环境中部署应用,并获得一项中国发明专利和一项美国发明专利的受理。
基金supported by the National Natural Science Foundation of China(Grant Nos.12302435 and 12221002)。
文摘Shock wave caused by a sudden release of high-energy,such as explosion and blast,usually affects a significant range of areas.The utilization of a uniform fine mesh to capture sharp shock wave and to obtain precise results is inefficient in terms of computational resource.This is particularly evident when large-scale fluid field simulations are conducted with significant differences in computational domain size.In this work,a variable-domain-size adaptive mesh enlargement(vAME)method is developed based on the proposed adaptive mesh enlargement(AME)method for modeling multi-explosives explosion problems.The vAME method reduces the division of numerous empty areas or unnecessary computational domains by adaptively suspending enlargement operation in one or two directions,rather than in all directions as in AME method.A series of numerical tests via AME and vAME with varying nonintegral enlargement ratios and different mesh numbers are simulated to verify the efficiency and order of accuracy.An estimate of speedup ratio is analyzed for further efficiency comparison.Several large-scale near-ground explosion experiments with single/multiple explosives are performed to analyze the shock wave superposition formed by the incident wave,reflected wave,and Mach wave.Additionally,the vAME method is employed to validate the accuracy,as well as to investigate the performance of the fluid field and shock wave propagation,considering explosive quantities ranging from 1 to 5 while maintaining a constant total mass.The results show a satisfactory correlation between the overpressure versus time curves for experiments and numerical simulations.The vAME method yields a competitive efficiency,increasing the computational speed to 3.0 and approximately 120,000 times in comparison to AME and the fully fine mesh method,respectively.It indicates that the vAME method reduces the computational cost with minimal impact on the results for such large-scale high-energy release problems with significant differences in computational domain size.
基金National Natural Science Foundation of China(62373102)Jiangsu Natural Science Foundation(BK20221455)Anhui Provincial Key Research and Development Project(2022i01020013)。
文摘Enhancing the stability and performance of practical control systems in the presence of nonlinearity,time delay,and uncertainty remains a significant challenge.Particularly,a class of strict-feedback nonlinear uncertain systems characterized by unknown control directions and time-varying input delay lacks comprehensive solutions.In this paper,we propose an observerbased adaptive tracking controller to address this gap.Neural networks are utilized to handle uncertainty,and a unique coordinate transformation is employed to untangle the coupling between input delay and unknown control directions.Subsequently,a new auxiliary signal counters the impact of time-varying input delay,while a Nussbaum function is introduced to solve the problem of unknown control directions.The leverage of an advanced dynamic surface control technique avoids the“complexity explosion”and reduces boundary layer errors.Synthesizing these techniques ensures that all the closed-loop signals are semi-globally uniformly ultimately bounded(SGUUB),and the tracking error converges to a small region around the origin by selecting suitable parameters.Simulation examples are provided to demonstrate the feasibility of the proposed approach.
基金funded by the National Natural Science Foundation of China (Grant No. 62471048)Open Fund Project of Beijing Key Laboratory of High Dynamic Navigation TechnologyKey Laboratory Fund Project of Modern Measurement and Control Technology, Ministry of Education
文摘In complex environments such as high dynamics and weak signals,a satellite signal compensation method based on prefabricated trajectory assistance and an improved adaptive Kalman filter is proposed for a 155 mm differential rotating rear-body control-guided projectile to address the situation of satellite signal flickering and loss in projectile navigation systems due to environmental limitations.First,establish the system state and measurement equation when receiving satellite signals normally.Second,a seven-degree-of-freedom external ballistic model is constructed,and the ideal trajectory output from the ballistic model is used to provide the virtual motion state of the projectile,which is input into a filter as a substitute observation when satellite signals are lost.Finally,an adaptive Kalman filter(AKF)is designed,the proposed adaptive Kalman filter can accurately adjust the estimation error covariance matrix and Kalman gain in real-time based on information covariance mismatch.The simulation results show that compared to the classical Kalman filter,it can reduce the average positioning error by more than 38.21%in the case of short-term and full-range loss of satellite signals,providing a new idea for the integrated navigation of projectiles with incomplete information under the condition of satellite signal loss.
基金supported by the National Natural Science Foundation of China under Grant 62301051.
文摘Range-azimuth imaging of ground targets via frequency-modulated continuous wave(FMCW)radar is crucial for effective target detection.However,when the pitch of the moving array constructed during motion exceeds the physical array aperture,azimuth ambiguity occurs,making range-azimuth imaging on a moving platform challenging.To address this issue,we theoretically analyze azimuth ambiguity generation in sparse motion arrays and propose a dual-aperture adaptive processing(DAAP)method for suppressing azimuth ambiguity.This method combines spatial multiple-input multiple-output(MIMO)arrays with sparse motion arrays to achieve high-resolution range-azimuth imaging.In addition,an adaptive QR decomposition denoising method for sparse array signals based on iterative low-rank matrix approximation(LRMA)and regularized QR is proposed to preprocess sparse motion array signals.Simulations and experiments show that on a two-transmitter-four-receiver array,the signal-to-noise ratio(SNR)of the sparse motion array signal after noise suppression via adaptive QR decomposition can exceed 0 dB,and the azimuth ambiguity signal ratio(AASR)can be reduced to below-20 dB.
基金supported by the Fundamental Research Funds for the Central Universities(2024JBZX038)the National Natural Science Foundation of China(62076023).
文摘With the rapid development of low-altitude economy and unmanned aerial vehicles (UAVs) deployment technology, aerial-ground collaborative delivery (AGCD) is emerging as a novel mode of last-mile delivery, where the vehicle and its onboard UAVs are utilized efficiently. Vehicles not only provide delivery services to customers but also function as mobile ware-houses and launch/recovery platforms for UAVs. This paper addresses the vehicle routing problem with UAVs considering time window and UAV multi-delivery (VRPU-TW&MD). A mixed integer linear programming (MILP) model is developed to mini-mize delivery costs while incorporating constraints related to UAV energy consumption. Subsequently, a micro-evolution aug-mented large neighborhood search (MEALNS) algorithm incor-porating adaptive large neighborhood search (ALNS) and micro-evolution mechanism is proposed. Numerical experiments demonstrate the effectiveness of both the model and algorithm in solving the VRPU-TW&MD. The impact of key parameters on delivery performance is explored by sensitivity analysis.
文摘As commercial drone delivery becomes increasingly popular,the extension of the vehicle routing problem with drones(VRPD)is emerging as an optimization problem of inter-ests.This paper studies a variant of VRPD in multi-trip and multi-drop(VRP-mmD).The problem aims at making schedules for the trucks and drones such that the total travel time is minimized.This paper formulate the problem with a mixed integer program-ming model and propose a two-phase algorithm,i.e.,a parallel route construction heuristic(PRCH)for the first phase and an adaptive neighbor searching heuristic(ANSH)for the second phase.The PRCH generates an initial solution by con-currently assigning as many nodes as possible to the truck–drone pair to progressively reduce the waiting time at the rendezvous node in the first phase.Then the ANSH improves the initial solution by adaptively exploring the neighborhoods in the second phase.Numerical tests on some benchmark data are conducted to verify the performance of the algorithm.The results show that the proposed algorithm can found better solu-tions than some state-of-the-art methods for all instances.More-over,an extensive analysis highlights the stability of the pro-posed algorithm.
