In this paper,a model reference adaptive control(MRAC)augmentation method of a linear controller is proposed for air-breathing hypersonic vehicle(AHV)during inlet unstart.With the development of hypersonic flight tech...In this paper,a model reference adaptive control(MRAC)augmentation method of a linear controller is proposed for air-breathing hypersonic vehicle(AHV)during inlet unstart.With the development of hypersonic flight technology,hypersonic vehicles have been gradually moving to the stage of weaponization.During the maneuvers,changes of attitude,Mach number and the back pressure can cause the inlet unstart phenomenon of scramjet.Inlet unstart causes significant changes in the aerodynamics of AHV,which may lead to deterioration of the tracking performance or instability of the control system.Therefore,we firstly establish the model of hypersonic vehicle considering inlet unstart,in which the changes of aerodynamics caused by inlet unstart is described as nonlinear uncertainty.Then,an MRAC augmentation method of a linear controller is proposed and the radial basis function(RBF)neural network is used to schedule the adaptive parameters of MRAC.Furthermore,the Lyapunov function is constructed to prove the stability of the proposed method.Finally,numerical simulations show that compared with the linear control method,the proposed method can stabilize the attitude of the hypersonic vehicle more quickly after the inlet unstart,which provides favorable conditions for inlet restart,thus verifying the effectiveness of the augmentation method proposed in the paper.展开更多
This paper focuses on self-supervised video representation learning.Most existing approaches follow the contrastive learning pipeline to construct positive and negative pairs by sampling different clips.However,this f...This paper focuses on self-supervised video representation learning.Most existing approaches follow the contrastive learning pipeline to construct positive and negative pairs by sampling different clips.However,this formulation tends to bias the static background and has difficulty establishing global temporal structures.The major reason is that the positive pairs,i.e.,different clips sampled from the same video,have limited temporal receptive fields,and usually share similar backgrounds but differ in motions.To address these problems,we propose a framework to jointly utilize local clips and global videos to learn from detailed region-level correspondence as well as general long-term temporal relations.Based on a set of designed controllable augmentations,we implement accurate appearance and motion pattern alignment through soft spatio-temporal region contrast.Our formulation avoids the low-level redundancy shortcut with an adversarial mutual information minimization objective to improve the generalization ability.Moreover,we introduce local-global temporal order dependency to further bridge the gap between clip-level and video-level representations for robust temporal modeling.Extensive experiments demonstrate that our framework is superior on three video benchmarks in action recognition and video retrieval,and captures more accurate temporal dynamics.展开更多
基金supported by the Foundation of Shanghai Aerospace Science and Technology(SAST2016077)。
文摘In this paper,a model reference adaptive control(MRAC)augmentation method of a linear controller is proposed for air-breathing hypersonic vehicle(AHV)during inlet unstart.With the development of hypersonic flight technology,hypersonic vehicles have been gradually moving to the stage of weaponization.During the maneuvers,changes of attitude,Mach number and the back pressure can cause the inlet unstart phenomenon of scramjet.Inlet unstart causes significant changes in the aerodynamics of AHV,which may lead to deterioration of the tracking performance or instability of the control system.Therefore,we firstly establish the model of hypersonic vehicle considering inlet unstart,in which the changes of aerodynamics caused by inlet unstart is described as nonlinear uncertainty.Then,an MRAC augmentation method of a linear controller is proposed and the radial basis function(RBF)neural network is used to schedule the adaptive parameters of MRAC.Furthermore,the Lyapunov function is constructed to prove the stability of the proposed method.Finally,numerical simulations show that compared with the linear control method,the proposed method can stabilize the attitude of the hypersonic vehicle more quickly after the inlet unstart,which provides favorable conditions for inlet restart,thus verifying the effectiveness of the augmentation method proposed in the paper.
基金supported in part by the National Natural Science Foundation of China(No.62325109,U21B2013).
文摘This paper focuses on self-supervised video representation learning.Most existing approaches follow the contrastive learning pipeline to construct positive and negative pairs by sampling different clips.However,this formulation tends to bias the static background and has difficulty establishing global temporal structures.The major reason is that the positive pairs,i.e.,different clips sampled from the same video,have limited temporal receptive fields,and usually share similar backgrounds but differ in motions.To address these problems,we propose a framework to jointly utilize local clips and global videos to learn from detailed region-level correspondence as well as general long-term temporal relations.Based on a set of designed controllable augmentations,we implement accurate appearance and motion pattern alignment through soft spatio-temporal region contrast.Our formulation avoids the low-level redundancy shortcut with an adversarial mutual information minimization objective to improve the generalization ability.Moreover,we introduce local-global temporal order dependency to further bridge the gap between clip-level and video-level representations for robust temporal modeling.Extensive experiments demonstrate that our framework is superior on three video benchmarks in action recognition and video retrieval,and captures more accurate temporal dynamics.
