The tracker based on Siamese neural network is currently a technical method with high accuracy in the tracking field. With the introduction of transformer in the visual tracking field, the attention mechanism has grad...
The tracker based on Siamese neural network is currently a technical method with high accuracy in the tracking field. With the introduction of transformer in the visual tracking field, the attention mechanism has gradually emerged in tracking tasks. However, due to the characteristics of attention operation, Transformer usually has slow convergence speed, and its pixel-level correlation discrimination in tracking is more likely to lead to overfitting, which is not conducive to long-term tracking. A brand new framework FAT was designed, which is the improvement of MixFormer. The operation for simultaneous feature extraction and target information integration in MixFormer is retained, and the Mixing block is introduced to suppress the background as much as possible before the information interaction. In addition, a new operation is designed: the result of region-level cross-correlation is used as a guidance to help the learning of pixel-level cross-correlation in attention, thereby accelerating the model convergence speed and enhancing the model generalization. Finally, a joint loss function is designed to further improve the accuracy of the model. Experiments show that the presented tracker achieves excellent performance on five benchmark datasets.
Given the dual robots play a key role for collaborative work. In this paper, an improved method is proposed to find the kinematic positive solution based on its D-H parameters, model the dual SCARA robot in MATLAB env...
Given the dual robots play a key role for collaborative work. In this paper, an improved method is proposed to find the kinematic positive solution based on its D-H parameters, model the dual SCARA robot in MATLAB environment, and solve the collaborative workspace point cloud map by Monte Carlo method. The proposed method extracts and plots the boundary curves of the collaboration space by improved the α-Shapes algorithm, and obtains the shape of the common region of the collaboration space of the dual SCARA robots and the limit position parameters in the collaboration space. It provides a theoretical basis for the research of path planning, trajectory planning, dynamics analysis, mechanism parameter optimization, self-avoidance, and anti-collision problems of the dual-robot collaborative system.
In this paper, the master-slave synchronization issue of chaotic Lur'e systems with time-varying-delay feedback control is investigated. Firstly, the synchronization problem of chaotic system is transformed into t...
In this paper, the master-slave synchronization issue of chaotic Lur'e systems with time-varying-delay feedback control is investigated. Firstly, the synchronization problem of chaotic system is transformed into the stability problem of chaotic synchronization error system, which is studied based on Lyapunov-Krasovskii functional(LKF) method. Secondly, a novel augmented LKF with more cross terms that related to time-varying delay is proposed. Based on the application of the relaxation integral inequality and the reciprocally convex matrix inequality, an improved synchronization criterion is derived by using the cubic function negative-determination lemma. Finally, a numerical simulation example demonstrates the effectiveness and advantages of the proposed methods.
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism(DETM). The DETM is designed to optimize packet ...
This paper explores the finite-time synchronization of a class of discrete-time nonlinear singularly perturbed complex networks using a dynamic event-triggered mechanism(DETM). The DETM is designed to optimize packet transmission, aiming to conserve network resources. By constructing a Lyapunov function considering singularly perturbed parameters(SPPs) and DETM information, a sufficient condition for the dynamics of synchronization error system to be finite-time stable is given. The parameters of the synchronization controller can be determined by solving a set of matrix inequalities. The effectiveness of the proposed controller is demonstrated through a numerical example.
With the remarkable development of the power grid, traditional manual maintenance modes cannot meet the requirement of efficiency and reliability. Robotic technology is applied for power line maintenance to bring down...
详细信息
A new Gaussian approximate(GA) filter for nonlinear systems with one-step randomly delayed measurement and correlated noise is proposed in this ***,a general framework of Gaussian filter is designed under Gaussian ass...
A new Gaussian approximate(GA) filter for nonlinear systems with one-step randomly delayed measurement and correlated noise is proposed in this ***,a general framework of Gaussian filter is designed under Gaussian assumption on the conditional ***,the implementation of Gaussian filter is transfomed into the approximation of the Gaussian weighted integral in the proposed ***,a new cubature Kalman filtering(CKF)algorithm is developed on the basis of the spherical-radial cubature *** efficiency and superiority of the proposed method are illustrated in the numerical examples.
In this paper, we investigate the simultaneous state and unknown input (SUI) filtering issue for a class of multi-sensor networked systems (MSNSs). The unknown input with no prior knowledge is introduced in the system...
详细信息
To enhance the estimation accuracy and dynamic performance of sensorless surface-mounted permanent magnet synchronous motor (SPMSM) drives, a sensorless control scheme based on generalized super-twisting observer (GST...
To enhance the estimation accuracy and dynamic performance of sensorless surface-mounted permanent magnet synchronous motor (SPMSM) drives, a sensorless control scheme based on generalized super-twisting observer (GSTO) and non-smooth controller is proposed. Firstly, a GSTO for back electromotive force (back-EMF) estimation is proposed. Compared with the conventional super-twisting observer, the GSTO has a faster convergence rate and stronger robustness due to the additional terms. Then a linear extended state observer (LESO) is adopted to estimate the position, speed, and lumped disturbance at the same time. Moreover, a non-smooth composite speed controller is designed by combining the disturbance feed-forward compensation. Compared with the conventional PI speed controller, the non-smooth controller has a shorter settling time and a better disturbance rejection ability. Finally, the effectiveness of the proposed method is verified by simulation results.
This paper uses the wave equation to explain the torsional motion of the drill-string system.S olving the wave equation with the D'Alembert method,a neutral time-delay model of the drill-string system is *** distu...
This paper uses the wave equation to explain the torsional motion of the drill-string system.S olving the wave equation with the D'Alembert method,a neutral time-delay model of the drill-string system is *** disturbance input,caused by the bit-rock interaction,is given consideration,and an equivalent-input-disturbance(EID) based controller is designed to mitigate the disturbance in the established *** the actual drilling procedure,the system input time-delay increases as the length of the drill columns *** the influence of system input time-delay in the drilling procedure is ignored,it will most likely lead to the drill-string system instability and cause serious *** essential contribution of this paper is the incorporation of input time-delay into the EID based control *** the system's input time-delay,the proposed model is more practical and has significant implications for stick-slip vibration assessment and control in drilling procedures.
暂无评论