Focusing on the issue that the computation of the optical flow has a weak robustness to the edge diffusion, noise and outliers of the visible flow image, an improved approach of physics-based optical flow computation is developed to enhance the robustness of the optical flow computation. The proposed algorithm introduces the anisotropic filter into the method of the physics-based optical flow to improve the robustness of edge optical flow, and integrates the penalty factor into this approach to reduce the effect of noise and outliers on the optical flow computation. Then, the energy function of optical flow is minimized by using variation principle to solve the Euler Lagrange equation. Finally, the velocity field is obtained by exploiting the iterative method.Simulation results show that, compared with the traditional Lucas-Kande, Horn-Schunck, Pyramid Lucas-Kande, and the physics-based optical flow computational methods, the proposed algorithm can significantly reduce the diffusion of optical flow on edges and in corners, improve the robustness to noise and outliers, and thus obtain a velocity field with higher robustness.

A new adaptive embedded Cubature Kalman Filter (CKF) algorithm based on the Sage-Husa algorithm is proposed for nonlinear target tracking when the distribution of noise is unknown. Firstly, the embedded cubature guidelines are used to improve the conventional Sage-Husa algorithm, thus a noise statistical estimator adaptive to the embedded CKF is obtained for estimating the statistical characteristics of the unknown noise and making correction. Then, a judgment mechanism is introduced into the algorithm to suppress the divergence of the filter. Finally, the result of the maneuvering target tracking simulation verifies the effectiveness of the proposed algorithm.

The airborne electro-optical sighting devices and seekers of the electro-optical guided weapons are the direct targets of the electro-optical countermeasure system in regional electronic air defence combat. Rapid and accurate target recognition and threat assessment form the basis for command decision-making and effective implementation of jamming. In this paper, the characteristics of the targets in the countermeasure are analyzed firstly. Considering the ambiguous battlefield situation, a target recognition method based on action reasoning is proposed to identify the target type and predict its next action. Finally, a multi-attribute decision-making threat assessment model is established, and used together with the effectiveness reduction analysis, to calculate target threat level. The simulation shows that, this method can make accurate judgment of target attribute and threat level and provide a basis for target assignment in jamming when there is no external intelligence support.

In the process of modal switching between different phases of flight, the parameters of the controller in different modes may change due to the sudden change of flight state, thus it is difficult to design a general controller to ensure the stability and smoothness of the switching process of both the modes. Direct switching between the two modal controllers may cause jumping of the control law parameters of the aircraft, and result in sudden change of the flight state, which will then affect the stability and smoothness of the modal switching process, make the system unstable and cause flight accident. In this paper, a fast double-power sliding mode switching controller based on inertia is designed taking the switching from the end climbing section to the initial cruise section as the example. Firstly, the sliding surface of the sliding mode before and after switching is used to get the sliding surface of the switching control law, and then the sliding mode switching control law is designed for the new sliding surface. Finally, the simulation results show that the fast double-power switching control method has a good tracking control effect.

A robust adaptive control scheme is designed for a class of pure-feedback nonlinear systems with non-affine functions possibly being discontinuous.The continuous and bounded conditions on non-affine functions are relaxedand the bounds of non-affine function are unknown continuous functions.Moreoverthe problem of computational complexity caused by the repeated differentiations of virtual control laws is eliminated by utilizing dynamic surface control technique.It is proved theoretically that all the signals in the closed-loop control system are semi-globally uniformly ultimately boundedand the system output can asymptotically converge to an arbitrarily small neighborhood of the equilibrium point by choosing the appropriate design parameters.Finally simulation examples are provided to demonstrate the effectiveness of the designed approach.

A global approximation convergence controller is designed for the free-floating space robot system, which solves the robustness problem of the system when there is initial state error or external disturbance.Firstly, the controller is designed based on the state space equation of the system, which isolates the output characteristics from control gains, and exhibits strong robustness against model uncertainties and external disturbance. Meanwhile, the control method completely avoids the high-complexity issue raised by backstepping-like control approaches.Then, it is proved theoretically that the proposed controller can ensure the robot system to track the desired path. Finally, the proposed controller and the robust compensation controller are used respectively for simulation analysis to a single-arm six-DOF space robot system.The simulation results confirm the superiority of the proposed controller.

This paper proposes a Synthetic Aperture Radar (SAR) target recognition method based on hierarchical decision fusion of complementary features. The Principal Component Analysis (PCA) feature, target peaks and contour are employed as basic features for target recognition, which are complementary with each other and can provide more comprehensive target descriptions.At the stage of decision fusionthe hierarchical decision fusion strategy is used. For target recognition, the first level employs the PCA feature, the second level uses the peaks, and the third level adopts the contour. When the former level recognizes the target with high reliability, the whole recognition process ends and no more classification is needed. Therefore, the efficiency of the SAR target recognition system is significantly improved by avoiding the unnecessary recognition process.To validate the effectiveness of the proposed method, experiments are conducted on public Moving and Stationary Target Acquisition and Recognition (MSTAR ) dataset.

The kernel ridge regression classifier used in adaptive color attribute tracker makes little use of information about the context, which may easily result in drifting in cases of fast target motion, partial occlusion or background clutter. To solve the problem, the interference peak position is determined at first by calculating the peak side lobe ratio of the response matrix. Then, the corresponding context constraints are introduced on the kernel ridge regression classifier to enhance the background utilization. Finally, a scale space filter is established for supporting multi-scale variation. The experimental results based on the Visual Tracker Benchmark dataset show that this method can achieve high tracking accuracy with a mild loss of frame rates.

This paper mainly studies the speed and robustness of the single-axis control system of the gyro stabilized platform. The sliding-mode variable structure is combined with neural network for controlling the gyro stabilized platform. Aiming at the chattering problem of traditional sliding mode variable structure method due to inaccurate model estimation, a neural network sliding mode variable structure control method is proposed using RBF neural network to approximate the unknown part of the model. Experiments show that using this method can effectively improve the systems speed and robustness.For the uniaxial control system of the gyro stabilized platform, the method can implement the angle tracking in 0.3 s, and the speed tracking in 0.6 s;the maximum error of angle tracking is 0.007 3, and the error converges to a neighbourhood of zero in 1 s.

When selecting data source from the multi-sensor information sources for the traditional optoeletronic tracking devices, the modes of manual control or switching according to the priority are usually used.In order to overcome the hysteresis, randomness and step effect of the switching mechanism, a multi-source data fusion control strategy is proposed. First, the adaptive variable tracking gate is calculated for planning the related areas, and the outliers are eliminated.Thenthe hierarchical integration structure is designed according to the related areas.Finally, the membership function is used to calculate the fusion weighting factor. The practical application has showed that, this integration control strategy can improve the tracking accuracy, reduce the overshoot, and enhance the robustness and reliability.

Aiming at the problem that the compression perceptual Hash algorithm based on discrete cosine transform is difficult to track the target in the case of illumination change, target deformation or local occlusion, this paper proposes an enhanced template-matching Hash algorithm based on interclass variance and discrete cosine transform. This algorithm is a target tracking algorithm with automatic template updating, which uses the interclass variance threshold segmentation and discrete cosine transform to extract the different characteristic information of the target, the rapidly intensified method of difference to generate the Hashing sequence for reducing the influence of illumination, and the drawer principle to shorten the time for Hamming distance comparison. Experiments were made by using this algorithm, the traditional Hash algorithm, and the compression perceptual Hash algorithm based on DCT for tracking the three public standard videos of David, Girl, and CarScale.The results show that, the algorithm can improve the success rate of the target tracking under illumination change, target deformation or local occlusion, has good robustness and satisfies the real-time tracking requirements.

The variable step-size LMS adaptive filtering algorithm adjusts the weights by constructing a step-size factor, so that the algorithm has faster convergence speed and smaller steady-state error. In order to further improve the performance of the algorithm, an improved variable step-size LMS adaptive algorithm based on S function is proposed. By using the curve characteristics of S function, the algorithm obtains the expression of the step-size factor by translation transformation of the function. According to the requirement of controllability and anti-interference ability, controllable parameters and error vector autocorrelation values are introduced to adjust the step-size factor, and the final model is obtained. An analysis is made to the influence of the values of these parameters on the step-size factor and the filtering performance. The simulation results show that, compared with the existing algorithms, the algorithm has better performance on convergence speed, steady-state error and anti-interference ability.

How to improve the acquisition speed and precision of satellite navigation positioning P-code is one of the hot issues in military receiving technology research. Based on the GPS P-code extended replication overlap (XFAST) acquisition algorithm and the mean acquisition algorithm, an acquisition algorithm based on time-frequency fusion is proposed.Through XFAST processing to the local sequence, the search area is expanded without affecting the degree of correlation. Then averaging is made to the processed local sequence and the received sequence, thus to reduce the amount of computation, and to achieve the purpose of improving the capture speed. The feasibility of the algorithm is analyzed, and the specific process of the algorithm is presented. The simulation results verify the effectiveness of the proposed method.

Embryonic circuit is a new kind of bio-inspired hardware with the capability of self-diagnosis and self-repair. It can effectively improve the reliability of electronic equipments in complex environment. Self-checking is the premise of embryonic circuit self-repairing and is an important factor of embryonic circuit reliability. It is also one of the key issues in embryonic circuit research. In this paper, the basic concept of self-checking technologies and the status of studies at home and abroad are introduced, and a classification is given. The features of different embryonic circuit self-checking technologies are studied, and the main problems are discussed.A prospect is given for future development of the technologies.

In order to improve the stability control accuracy of airborne electro-optical platform for Unmanned Aerial Vehicles (UAVs), a kind of cascade stability control method based on double speed-loop was designed in consideration of the low disturbance rejection performance of single speed-loop control.The double speed-loop structure was constructed by using the velocity gyroscope to form the inner loop, and the optoelectronic encoder to form the outer loop. Theoretic analysis and experiments were carried to the single/double-loop structures on their disturbance rejection performance, robustness and dynamic response performance.The results showed that the double speed-loop structure is more robust and has better performance on disturbance rejection and dynamic response.The method can effectively improve the stability control accuracy of the electro-optical platform mounted on UAVs.

Aiming at the problems of Qball2 quad-rotor UAV of under-actuation, uncertain parameters, and strong coupling, a control method is proposed combining Tracking Differentiator (TD) with Linear Active Disturbance Rejection Control (LADRC). The LADRC is composed of two parts:Linear Combination (PD) and Linear Expansion State Observer (LESO).LESO makes real-time observation and dynamic compensation for the internal and external disturbances, thus the complex Qball2 system can be simplified to a series of integral forms. The LADRC parameter tuning is simple, and the introduction of TD makes the system fast and robust. The simulation and Qball2 experiment results show that the designed controller can satisfy the requirements of the system on dynamic performance and steady-state performance.

A novel Fault Tolerant Control (FTC) scheme is developed based on Adaptive Dynamic Programming (ADP) for a class of affine nonlinear systems with actuator failures. The estimated actuator failure from a fault observer is utilized to construct an improved performance index function, so the FTC problem can be transformed into an optimal control problem. By using Policy Iteration(PI)algorithm, the Hamilton-Jacobi-Bellman equation can be solved by constructing a critic neural network. Then, the approximate optimal controller can be derived directly. The closed-loop system is guaranteed to be asymptotically stable via the Lyapunov stability theorem. Finally, an example is presented to demonstrate the effectiveness of the proposed method.

Due to the characteristics of multi-input/multi-output, strong coupling, nonlinearity and under-actuation, the quad-rotor UAVs are sensitive to the external disturbances. To solve the problem, a control system based on Integral Backstepping is proposed. Firstly, the dynamic model of quad-rotor aircraft is established. Then, a double closed-loop control structure PID controller is designed with the outer-loop of the position control loop and the inner-loop of the attitude control loop, and a controller based on Integral Backstepping method is also designed, which is proved by Lyapunov stability theory.Finally, a simulation analysis is made to the two kinds of controllers with Matlab/Simulink. The results show that the controller model using Integral Backstepping design is obviously superior to the classical PID algorithm on control precision, regulating time and anti-disturbance performance.

In order to monitor the six-DOF deformation via monocular vision, a circle-pair tracking algorithm based on direction-prediction, a contour extraction algorithm and an occlusion identification method based on watershed, were studied, and a monitoring system based on embedded platform was designed and developed.The system consists of a target module, a visual collecting module, a CPU module and the result display interfaces. There are 5 pairs of circular white marks of the same size on the target module, which fix the target on the outside surface of an object to be monitored. Through tracking the relative positions of the 5 pairs of marks, the displacements and rotation angles of the object to be monitored in Cartesian rectangular coordinate can be calculated.

In the Measurement Assurance Program (MAP), due to the high-frequency data acquisition of the measurement and control equipment, the data may present dependency on timing sequence, and can hardly meet the data independence requirements of the traditional control chart.It is proposed to simulate the auto-regression model AR (p) with the observation data in the constrained state according to the timing sequence principle, the residual of the observation value is calculated, which is independently and identically distributed, and the control chart is set up. To further illustrate the validity of the residual control chart, real data is used to obtain the residual control chart and then it is compared with the Shewhart control chart.The results show that the occurrence of the false alarm of the residual control chart is significantly reduced.

Aiming at the problems that the main structure of a mini-optical device for spatial use is too heavy, its ground gravity deformation is severe and the base frequency is too low, an optimization mathematical model was established with the minimum mass and the RMS of random acceleration response as the objectives, and the fundamental frequency and the deformation as the constraint conditions. The topology optimization design of the main structure was carried out. An engineering analysis was made to the optimized main structure. The results showed that the optimized main structure of the mini-optical device has lighter mass, higher fundamental frequency, less deformation and random vibration response.Mechanics experiments were carried out to examine the performance of the main structure. The detection results accord with the overall index, which proved that the optimization method is effective and feasible.

Aiming at the problems that the traditional DS/FH signal source is less intelligent, inconvenient to carry and inefficient, a DS/FH signal source with small volume, low power consumption and high precision was designed based on FPGA+AD9915. The system software was designed with Verilog language by using Quartus Ⅱ software development platform. The FPGA chip was used to realize the baseband signal processing, frequency hopping code generating and intermediate frequency modulating, and the control of AD9915 and digital-analog conversion chip was also realized through FPGA. Finally, the DS/FH signal source was generated by mixing frequency. The test results show that the system has better performance and can meet the test requirements on frequency hopping rate and frequency interval.