Search

A driving controller design which can drive 512×512 resistor array at 200Hz frame rate is proposed. In view of the strict requirements of MOS resistor array drive controller for real-time performance, FPGA, optical fiber transmission card and PC104 technologies are adopted in the design, which studies and solves the two key problems of how to transmit image data at high speed and how to quickly generate multi-channel analog drive signals, and leaves expansion space for the drive control of larger resistor array in the future.

In this study, the authors study the distributed finite-time attitude dynamic tracking control problem for multiple rigid spacecraft systems with a single dynamic leader under directed graphs. By constructing a sliding-mode estimator for each follower, they remove the requirement of the angular acceleration information. Based upon a sliding-mode observation vector, the distributed input torques are designed to guarantee that not only the attitude but also the angular velocity of the followers will converge to the leader's. Necessary and sufficient criteria are established for the proposed protocol. Finally, the theoretical result is illustrated by a numerical simulation.

This paper provides an overview of attitude coordination problems of multi-rigid-body with the goal of promoting research in this area. Theoretical results regarding consensus seeking with different system models, different communication topologies, different control goals and different techniques are summarized. And applications of consensus protocols to multirigid- body coordination are investigated. Finally, some future research directions and open problems are also proposed.

Small target detection in infrared video sequence is a challenging problem. In this paper, a collaborative structured sparse coding (SSC) model which incorporates the L 1 , 2 and L 2 , 1 regularization terms is proposed. The Alternating Direction Method of Multiplier (ADMM) is developed to solve this model. Further, online dictionary learning is embedded into the model and temporal information is incorporated to eliminate the clutters and noises. Extensive synthetic and real data experiments show that our method obtains better detection performance than baseline methods and state-of-art infrared-patch-image (IPI) model.

The Airplane 3D geometric model is established with unstructured mesh dividing. The numerical simulation of the airplane outflow field is carried out based on the commercial CFD software, the temperature distribution data of aircraft surface is obtained, the radiation process of the tail nozzle and plume radiation is modeled and calculated. The calculation model of infrared radiation characteristic of airframe is established based on Reverse Monte Carlo method, the infrared irradiance of plume is calculated based on radiative transfer equation, The infrared radiation characteristics of the aircraft is obtained, the infrared image of airplane is obtained by method of gray level transformation.

The Newton–Kantorovich/Chebyshev pseudospectral method is systematically investigated to solve perturbed astrodynamic two-point boundary-value problems. Because many unperturbed astrodynamic problems have been solved and the perturbations are relatively small, the unperturbed solutions can serve as good initial approximations. The differential equations describing the perturbed problem are linearized about the nominal (approximate) solution, and the pseudospectral method is used to solve the linearized differential equations. By successively updating the nominal solution with the Newton–Kantorovich approach, the linearized solving procedure is iterated to achieve high-order precision. Because no numerical integration is needed and only linear algebraic equations are solved, the method is computationally efficient. The perturbed Lambert’s problem, the relative motion of spacecraft, and the optimal control of spacecraft formation flight are solved. High-precision solutions are obtained with only a few itera...

The hyperspectral unmixing technique is of great significance for hyperspectral data analysis. The non-negative matrix factorization (NMF)-based method is one of the most popular unmixing methods. However, the sparse constraints commonly used in NMF methods are not differentiable at zero, which affects the stability and accuracy of the results. In addition, sparse constraints are supposed to have endmember departure limitations. An improved sparse NMF algorithm is proposed. First, a biased L1 / 2 sparse constraint is used to effectively explore the sparsity of abundances, which can also have a sparse effect under the condition of the abundance sum-to-one constraint. Second, to alleviate the endmember departure effect, a minimum energy difference constraint is introduced. Simulated and real-data experiments verify the advantages of the proposed method against state-of-the-art sparse NMF methods.

Remote sensing image classification is a very challenging problem and covariance descriptor can be introduced in the feature extraction process for remote sensing image. However, covariance descriptor lies in non-Euclidean manifold, and conventional extreme learning machine (ELM) cannot effectively deal with this problem. In this paper, we propose an improved ELM framework incorporating the collaborative coding to tackle the covariance descriptor classification problem. First, a new ELM-guided dictionary learning and coding model is proposed to represent the covariance descriptor. Then the iterative optimization algorithm is developed to solve the model. By evaluating the proposed approach on the public dataset, we show the effectiveness of the proposed strategy.

Pedestrian detection in thermal infrared image is a challenging and hot topic. In this paper, a novel and robust pedestrian detection method based on Binarized Normed Gradients (BING) and Extreme Learning Machine (ELM) is proposed. The candidates are firstly generated using BING, and then ELM is used to classify a candidate to be pedestrian or not. Experiment results indicate that our method can obtain better performance than traditional Support Vector Machine (SVM).

Orbit maintenance problems are solved by the synthesis of the traditional penalty function method and evolution algorithms (EAs). A general method is presented for the n-time impulse problem of orbit maintenance. The algorithm is proved to be effective according to a simulation example which adjusts the semi-major axis with 3,4,5 impulse times.