伺服与扰动抑制是时滞积分系统最基本的控制问题,对其进行控制难度较大。文中提出一种基于直接综合法和多主导极点配置法的微分先行PID(Proportional-Integral-Derivative)整定方法,这种方法通过比较串联滤波器与时滞积分被控对象组成...伺服与扰动抑制是时滞积分系统最基本的控制问题,对其进行控制难度较大。文中提出一种基于直接综合法和多主导极点配置法的微分先行PID(Proportional-Integral-Derivative)整定方法,这种方法通过比较串联滤波器与时滞积分被控对象组成的特征方程与实际期望的特征方程的系数,将三阶主导极点置于-1/λ处,并将二阶非主导极点置于-5/λ处(λ为调整参数),从而获得期望的特征方程。以实现期望的鲁棒性方式获得设计的控制器参数,通过选择不同的调优参数获取相应的Ms(Maximum sensitivity)值,在参数具有标称性的限定条件下拟合出关于Ms和调优参数的关系曲线,给出整定规则的解析形式。PIPTD(Pure Integral Plus Time Delay system)、DIPTD(Double Integral Plus Time Delay system)和FOPTDI(First-Order Plus Time Delay Integral System)系统的仿真结果表明,IAE(Integral Absolute Error)指标平均可降低35.79%,TV(Total Variation)指标平均可降低18.97%。展开更多
Video prediction is the problem of generating future frames by exploiting the spatiotemporal correlation from the past frame sequence.It is one of the crucial issues in computer vision and has many real-world applicat...Video prediction is the problem of generating future frames by exploiting the spatiotemporal correlation from the past frame sequence.It is one of the crucial issues in computer vision and has many real-world applications,mainly focused on predicting future scenarios to avoid undesirable outcomes.However,modeling future image content and object is challenging due to the dynamic evolution and complexity of the scene,such as occlusions,camera movements,delay and illumination.Direct frame synthesis or optical-flow estimation are common approaches used by researchers.However,researchers mainly focused on video prediction using one of the approaches.Both methods have limitations,such as direct frame synthesis,usually face blurry prediction due to complex pixel distributions in the scene,and optical-flow estimation,usually produce artifacts due to large object displacements or obstructions in the clip.In this paper,we constructed a deep neural network Frame Prediction Network(FPNet-OF)with multiplebranch inputs(optical flow and original frame)to predict the future video frame by adaptively fusing the future object-motion with the future frame generator.The key idea is to jointly optimize direct RGB frame synthesis and dense optical flow estimation to generate a superior video prediction network.Using various real-world datasets,we experimentally verify that our proposed framework can produce high-level video frame compared to other state-ofthe-art framework.展开更多
The transmission modes of multi-hop and broadcasting for Wireless Sensor Networks(WSN)often make random and unknown transmission delays appear,so multisensor data fusion based ondelayed systems attracts intense attent...The transmission modes of multi-hop and broadcasting for Wireless Sensor Networks(WSN)often make random and unknown transmission delays appear,so multisensor data fusion based ondelayed systems attracts intense attention from lots of researchers.The existing achievements for thedelayed fusion all focus on Out-Of-Sequence Measurements(OOSM)problem which has many dis-advantages such as high communication cost,low computational efficiency,huge computational com-plexity and storage requirement,bad real-time performance and so on.In order to overcome theseproblems occurred in the OOSM fusion,the Out-Of-Sequence Estimates(OOSE)are considered tosolve the delayed fusion for the first time.Different from OOSM which belongs to the centralized fusion,the OOSE scheme transmits local estimates from local sensors to the central processor and is thus thedistributed fusion;thereby,the OOSE fusion can not only avoid the problems suffered in the OOSMfusion but also make the design of fusion algorithm highly simple and easy.Accordingly,a novel optimallinear recursive prediction weighted fusion method is proposed for one-step OOSE problem in this letter.As a tradeoff,its fusion accuracy is slightly lower than that of the OOSM method because the currentOOSM fusion is a smooth estimate and OOSE gets a prediction estimate.But,the smooth result of theOOSE problem also has good fusion accuracy.Performance analysis and computer simulation show thatthe total performance of the proposed one-step OOSE fusion algorithm is better than the current one-step OOSM fusion in the practical tracking systems.展开更多
In the literature （Tan and Wang, 2010）, Tan and Wang investigated the convergence of the split-step backward Euler （SSBE） method for linear stochastic delay integro-differential equations （SDIDEs） and proved the...In the literature （Tan and Wang, 2010）, Tan and Wang investigated the convergence of the split-step backward Euler （SSBE） method for linear stochastic delay integro-differential equations （SDIDEs） and proved the mean-square stability of SSBE method under some condition. Unfortu- nately, the main result of stability derived by the condition is somewhat restrictive to be applied for practical application. This paper improves the corresponding results. The authors not only prove the mean-square stability of the numerical method but also prove the general mean-square stability of the numerical method. Furthermore, an example is given to illustrate the theory.展开更多
In this paper,we investigate the numerical performance of a family of P-stable two-step Maruyama schemes in mean-square sense for stochastic differential equations with time delay proposed in[8,10]for a certain class ...In this paper,we investigate the numerical performance of a family of P-stable two-step Maruyama schemes in mean-square sense for stochastic differential equations with time delay proposed in[8,10]for a certain class of nonlinear stochastic delay differential equations with multiplicative white noises.We also test the convergence of one of the schemes for a time-delayed Burgers’equation with an additive white noise.Numerical results show that this family of two-step Maruyama methods exhibit similar stability for nonlinear equations as that for linear equations.展开更多
基金supported by National Natural Science Foundation of China(61403149,61573298)Natural Science Foundation of Fujian Province(2015J01261,2016J05165)Foundation of Huaqiao University(Z14Y0002)
文摘伺服与扰动抑制是时滞积分系统最基本的控制问题,对其进行控制难度较大。文中提出一种基于直接综合法和多主导极点配置法的微分先行PID(Proportional-Integral-Derivative)整定方法,这种方法通过比较串联滤波器与时滞积分被控对象组成的特征方程与实际期望的特征方程的系数,将三阶主导极点置于-1/λ处,并将二阶非主导极点置于-5/λ处(λ为调整参数),从而获得期望的特征方程。以实现期望的鲁棒性方式获得设计的控制器参数,通过选择不同的调优参数获取相应的Ms(Maximum sensitivity)值,在参数具有标称性的限定条件下拟合出关于Ms和调优参数的关系曲线,给出整定规则的解析形式。PIPTD(Pure Integral Plus Time Delay system)、DIPTD(Double Integral Plus Time Delay system)和FOPTDI(First-Order Plus Time Delay Integral System)系统的仿真结果表明,IAE(Integral Absolute Error)指标平均可降低35.79%,TV(Total Variation)指标平均可降低18.97%。
基金supported by Incheon NationalUniversity Research Grant in 2017.
文摘Video prediction is the problem of generating future frames by exploiting the spatiotemporal correlation from the past frame sequence.It is one of the crucial issues in computer vision and has many real-world applications,mainly focused on predicting future scenarios to avoid undesirable outcomes.However,modeling future image content and object is challenging due to the dynamic evolution and complexity of the scene,such as occlusions,camera movements,delay and illumination.Direct frame synthesis or optical-flow estimation are common approaches used by researchers.However,researchers mainly focused on video prediction using one of the approaches.Both methods have limitations,such as direct frame synthesis,usually face blurry prediction due to complex pixel distributions in the scene,and optical-flow estimation,usually produce artifacts due to large object displacements or obstructions in the clip.In this paper,we constructed a deep neural network Frame Prediction Network(FPNet-OF)with multiplebranch inputs(optical flow and original frame)to predict the future video frame by adaptively fusing the future object-motion with the future frame generator.The key idea is to jointly optimize direct RGB frame synthesis and dense optical flow estimation to generate a superior video prediction network.Using various real-world datasets,we experimentally verify that our proposed framework can produce high-level video frame compared to other state-ofthe-art framework.
基金the National Natural Science Foundation of China(No.60434020,No.60572051)International Coop-erative Project Foundation(No.0446650006)Ministryof Education Science Foundation of China(No.2050 92).
文摘The transmission modes of multi-hop and broadcasting for Wireless Sensor Networks(WSN)often make random and unknown transmission delays appear,so multisensor data fusion based ondelayed systems attracts intense attention from lots of researchers.The existing achievements for thedelayed fusion all focus on Out-Of-Sequence Measurements(OOSM)problem which has many dis-advantages such as high communication cost,low computational efficiency,huge computational com-plexity and storage requirement,bad real-time performance and so on.In order to overcome theseproblems occurred in the OOSM fusion,the Out-Of-Sequence Estimates(OOSE)are considered tosolve the delayed fusion for the first time.Different from OOSM which belongs to the centralized fusion,the OOSE scheme transmits local estimates from local sensors to the central processor and is thus thedistributed fusion;thereby,the OOSE fusion can not only avoid the problems suffered in the OOSMfusion but also make the design of fusion algorithm highly simple and easy.Accordingly,a novel optimallinear recursive prediction weighted fusion method is proposed for one-step OOSE problem in this letter.As a tradeoff,its fusion accuracy is slightly lower than that of the OOSM method because the currentOOSM fusion is a smooth estimate and OOSE gets a prediction estimate.But,the smooth result of theOOSE problem also has good fusion accuracy.Performance analysis and computer simulation show thatthe total performance of the proposed one-step OOSE fusion algorithm is better than the current one-step OOSM fusion in the practical tracking systems.
基金supported by the Fundamental Research Funds for the Central Universities under Grant No. 2012089:31541111213China Postdoctoral Science Foundation Funded Project under Grant No.2012M511615the State Key Program of National Natural Science of China under Grant No.61134012
文摘In the literature （Tan and Wang, 2010）, Tan and Wang investigated the convergence of the split-step backward Euler （SSBE） method for linear stochastic delay integro-differential equations （SDIDEs） and proved the mean-square stability of SSBE method under some condition. Unfortu- nately, the main result of stability derived by the condition is somewhat restrictive to be applied for practical application. This paper improves the corresponding results. The authors not only prove the mean-square stability of the numerical method but also prove the general mean-square stability of the numerical method. Furthermore, an example is given to illustrate the theory.
基金This work was supported by the NSF of China(No.10901036)and AIRFORCE MURI.The authors thank the referees for their helpful suggestions for improving the paper.The first author also would like to thank Professor George Em Karniadakis for his hospitality when she was visiting Division of Applied Mathematics at Brown University.
文摘In this paper,we investigate the numerical performance of a family of P-stable two-step Maruyama schemes in mean-square sense for stochastic differential equations with time delay proposed in[8,10]for a certain class of nonlinear stochastic delay differential equations with multiplicative white noises.We also test the convergence of one of the schemes for a time-delayed Burgers’equation with an additive white noise.Numerical results show that this family of two-step Maruyama methods exhibit similar stability for nonlinear equations as that for linear equations.
