Search

Underwater optical image play a significant role in the exploration ocean. However there are diverse causes of distorted underwater optical scenes, such as light refraction, absorption, scattering and so on. Images enhancement is indispensable in low-level and high-level underwater vision task. Therefore, a novel method based on Generative Adversarial Networks is presented in this paper, which is able to recover lost information for underwater distorted images. No matter from color, detail or texture, the underwater images reconstructed by our approach been greatly improved. In addition, a novel solution is provided for lacking paired dataset which is needed for network train. At last, many qualitative and quantitative experiments are carried out, which can demonstrate that our approach proposed in this paper is robust and effective.

Surface defect detection has been spotlighted in the product quality control. There are lots of methods focused on the handcrafted optical features and have worked well under specified conditions. However, effectively detecting defects in products is nontrivial. Among the challenge is the complexity of surface defect, such as micro defect with noise, at vastly different scales. In order tackle these problems, we propose a feature fusion network using global context block for surface defect detection. A pipeline is presented that evaluates defect images with 300×300 resolution. In the framework, the global context block is relined, which fuses information effectively between different feature maps. Experimental results on steel defect datasets prove that our approach yields scores of map > 0.6 for all surface defects and provides a remarkably fast test speed, at ~ 20 frames per second.

In this paper, aiming at the discontinuous ignition reentry optimization problem of powered hypersonic vehicle, the reentry trajectory characteristics at different ignition time and ignition duration are analyzed by multiphase Gauss pseudospectral method. Given the attack angle and piecewise linear function described bank angle profiles, the problem of discontinuous ignition for powered reentry trajectory optimization is transformed into a parameter planning problem of ignition time, ignition duration, bank angle parameters and total flight time. By solving the nonlinear multi-constraint parameter optimization problem, the optimal ignition time, ignition duration and bank angle are obtained. Zero value bank angle and real-time calculation bank angle simulation examples are designed to optimize bank angle and ignition commands, respectively. The simulation results show that the discontinuous ignition reentry trajectory optimization problem of powered hypersonic vehicle can be solved by parameterized bank angle profile.

Accuracy of urban positioning depends on the quality of signals received by a Global Navigation Satellite System (GNSS) receiver that can be a combination of light of sight, reflection and diffraction signals. The more light of sight signals are received, the more accurate positioning can be achieved. To evaluate signal propagation thus availability of GNSS satellites in urban environments, a ray-tracing based modelling technique is proposed that can predict the light of sight, reflection and diffraction signals together with 3D city model. The developed techniques are validated against in-situ measurements in urban areas with different scenarios.

In order to improve the uniformity of magnetization and excitation efficiency of the two-dimensional rotational single sheet testers (RSST), In this paper, the simulation data of RSST with different shapes of sample are calculated by finite element method (FEM), and the magnetization uniformity of the surface of different shape samples is compared. A stereoscopic structure is designed by selecting Square Rotational Single Sheet Tester (SRSST) as the optimization object. The coupling effect of the orthogonal magnetic circuit and the influence of stray magnetic field between the magnetic poles are reduced by adding the air gap between the magnetic poles and the sample, the same effects can be caused by adding the shields up and down the sample. The sample magnetization uniformity of the SRSST is improved and the excitation power of the SRSST is reduced.

The satellite visibility of global navigation satellite systems in urban areas is investigated with electromagnetic ray-based methods and 3D city models. The satellite position is computed with satellite orbital parameters and the SGP4 standard mathematical model of satellite orbits. The obstruction, reflection and diffraction of satellite signals due to the presence of buildings are taken into account in the prediction of satellite visibility, with the satellite signal propagation paths being determined by using the ray tracing and 3D city models. In some cases where it is hard for a receiver to detect a satellite due to too low signal level, there do exist predicted signal propagation paths from the satellite to the receiver via reflection or diffraction. The received signal level is also estimated with the electromagnetic ray-based methods to enable the predicted satellite visibility to be correlated with that observed.

In view of the structure and characteristics of FFT algorithm, this paper designed an optimized 1024-point and 16-bit fixed-point FFT and it was implemented based on FPGA. This design adopted the algorithm of radix-2 with decimation in time(DIT), according to system performance and resource consumption, pipelined structure was selected. On the Altera-based CYCLONE II hardware platform, the algorithm was implemented using Verilog HDL language. After the FFT algorithm was transplanted to the Ion Mobility Spectrometer (IMS), air reactive ion and Dimethyl Methyl Phosphate (DMMP) experiments were carried out. The experimental results show that the signal-to-noise ratio (SNR) and the root mean square error (RMSE) of the IMS signal have been significantly improved.

In this paper, path planning of automated guided vehicle (AGV) in complex environment was investigated. A heterogeneous feature ant colony optimization algorithm based on effective vertexes of obstacles (EV-HFACO) is proposed to solve the problem of poor convergence and local optimum. Firstly, the grid method based on effective vertexes of obstacles played a very positive role as the method of environment modeling. Then, the different state rules which incorporate perspective-taking ability were designed. In addition, a new pheromone update mechanism was employed to balance the rapidity and randomness. Finally, some numerical results of simulation indicated the effectiveness of the proposed algorithms which could outperform that of popular SoSACO-v2” Sci-ACO and HHACO algorithms.

Millimeter-wave communication and Massive MIMO are key candidate technologies for 5G and hybrid beamforming is a promising solution to get better trade-off between system performance and hardware complexity. User selection and analog beam selection are two key problems in multi-user Millimeter-wave systems. We focus on the principle of orthogonality and energy maximization, a group-based user scheduling scheme is proposed with agglomerative clustering and beam steering beam selection. Numerical results show that agglomerative grouping scheduling can achieve considerable sum rates with affordable computation complexity.

The paper proposes an event-triggered model predictive control (MPC) scheme for formation control of multiple unmanned aerial vehicles (UAVs). A distributed MPC framework is designed in which each UAV only shares the information with its neighbors, and the obtained local finite-horizon optimal control problem (FHOCP) can be solved by swarm intelligent optimization algorithm. An event-triggered mechanism is proposed to reduce the computation burden for the distributed MPC scheme, which takes into consideration of the predictive state errors. Further, numerical simulations show that the proposed event-triggered MPC is more computationally efficient to achieve formation control of multiple UAVs in comparison with the tradition distributed MPC scheme.

This paper aims to build up a control platform for a small quadrotor to complete the design work of flight control system, so that the quadrotor can move freely in three-dimensional space. Firstly, a reference coordinate system and the motion control model of quadrotor are established. The PID method of flight control system of quadrotor is chosen, and the attitude and position control of the aircraft are modeled and simulated in MATLAB. Secondly, the detection data of gyroscope and accelerometer are calculated preliminarily and then processed through filtering algorithm by using STM32 micro-controller. The results showed that the attitude and position loop are controlled smoothly under small-angle action, and the control law is verified. The design of the function modules of the quadrotor’s attitude detection and flight control is completed. Therefore, the content of the research is generally correct and feasible. From these results, it is concluded that the hardware platform of quadrotor should be built up primarily, and the design of system software should be completed after the integration of each module. The attitude control and fixed altitude flight control of quadrotor and tracking flight mission can be achieved preliminarily in our control platform.

This paper uses five six-rotor UAVs as research objects, and focuses on the research of ad-hoc network, which is based on multi-rotor UAV’s organization, maintain, reconstruction and obstacle avoidance. It takes flight test of the multi-rotor UAV’s organization, which is based on ad-hoc strategy and the flight of anti-collision formation control system. The test results show that the formation of centralized control system has a stable and reliable way about communication and multi-vehicle formation. The latter is based on the optimization of single vehicle successfully tested formation construction, maintain, reconstruction and the anti-collision function in transforming formation’s process. It has important reference value in engineering implementation for multi-rotor UAV’s formation technology.

This paper focuses on the flight mode transition controller design of a small scaled V-44 which is a quad-tilt-rotor unmanned aerial vehicle combining the high-speed cruise and vertical take-off and landing capabilities. But there are some significant changes in aerodynamic characteristics during the mode transition which are difficult to measure. A model is established based on the momentum theory and Leaf integral method. To keep the height during the mode transition, a gain-scheduling controller is designed. Then the parameters of the controller is optimized by the Grey wolf optimization algorithm (GWO). Numerical simulations are conducted and the results show the excellent performance of the GWO-based gain-scheduling controller.

In this paper, a new active disturbance rejection control (ADRC) scheme based on swarm intelligent method is proposed for quadrotors to achieve position tracking and attitude stabilization. First, the finite-time convergent extended state observer (FTCESO) is designed to enhance the performance of ADRC controller. Then, the chaotic grey wolf optimization (CGWO) algorithm is developed with chaos initialization and chaos search to obtain the optimal parameters of attitude and position controllers. Numerical simulations are presented to demonstrate the effectiveness of the CGWO-based ADRC scheme.

2.4 GHz band is going to be congested in the aircraft, hence wireless systems using this band are highly susceptible to interference. The deployment of a wireless transmitting system using another frequency band in an aircraft cabin is investigated. The system operates at 60 GHz band with 2.16 GHz bandwidth for each channel and thus it is capable of very high throughput. Directional transmitting antennas are designed and beamforming technologies are employed, which further reduces the inference among neighboring networks.

In this work, a compact antenna which can be used for Ultra-Wide-Band (UWB) indoor localization is designed and simulated. Its performances are analyzed.

This paper focuses on the controller design of a hybrid autogyro which is composed of a Super-G and wings of Selig S3010. A traditional PID controller is first designed and then optimized through the particle swarm optimization (PSO) algorithm. Since the accurate dynamics model of the hybrid model is difficult to establish, the fuzzy and PSO-based PID controller is proposed which does not require a high-fidelity model. Numerical simulations are conducted and the results show the adaptability and fastness of the fuzzy-based PID controller.

This paper presents an optimization algorithm for trajectory planning of multiple unmanned aerial vehicles (UAVs). The goal is to generate three-dimensional trajectories satisfying the initial and terminal state constraints. The framework of cooperative trajectory planning is developed based on receding horizon optimization (RHO). The solution of trajectory optimization is obtained by using the convex optimization method, which serves as a convenient tool for dealing with equality and inequality constraints. The obstacle avoidance is also achieved by transforming the non-convex constraints to convex constraints. The effectiveness of the trajectory planning algorithm is tested by numerical simulations.

An improved ant colony algorithm is proposed in the actual storage environment on the basis of safety and stability. Keeping the omni-directional mobile vehicle away from obstacles is concerned first in order to reduce the risk of collision in consideration of the car size and the diversity of the obstacle species. Then, redundant points are simplified in the path and fewer inflection path is filtered based on grip map of environment so as to smooth the path. More flexible path selection strategy is selected, which further reduces the number of inflection points and deflection angle in order to get a more smooth path. What's more, some simulations are also given to illustrate the effectiveness of the improved algorithm. In addition, an evaluation function is constructed to compare the performance of the algorithm and results of arithmetic shows the improved algorithm is more security and stability.

This paper first briefly surveys the near-field scanning for antenna measurements, then focuses on the near-field scanning for electromagnetic compatibility (EMC) applications. Key theories and practices are covered, and some latest development and open issues regarding the near-field scanning for EMC applications are finally presented.

In this paper the triangular close formation flight control problem is addressed based on Lyapunov theory. Firstly, the three-dimensional nonlinear mathematical model of the formation flight control based on “leader-wing” mode was established. Secondly, both the dynamic and kinematic characteristics of the formation under the effect of aerodynamic coupling were analysed. Finally, the formation-hold autopilot for the follower aircraft is designed such that the formation's geometry is maintained according to the leader's maneuvering. Simulation showed that the system is able to form and maintain a stable triangular close formation rapidly with small formation error and has a good anti-interference performance.

Flight control system plays an important role in the process of missile designing. The quality of flight control system will affect the missile's performance significantly. In this article, we design a flight control system using variable structure control theory. Firstly, we establish a mathematical model of the missile and the basic content of variable structure control theory is introduced. Next chapter are mainly aiming at designing the control system of the missile's normal overload. Consequently, the linearized missile's normal overload is designed by using the mathematical model in chapter two. Then the variable structure controller is designed. In addition, we use the variable structure controller to optimize the linear normal load control system. The numerical simulation of two different control methods are given. And the result illustrate that the robustness of the control system based on variable structure theory are better.

In recent years, quadrotor unmanned aerial vehicles have played an important role in many aspects. Therefore, people put higher demands on the robustness of the quadrotor UAV's control. In this paper, a control algorithm based on signal compensation is proposed. Compared with the PID algorithm, this algorithm has better robustness to external disturbance and modeling uncertainty. Moreover, the controller of this algorithm is relatively simple and easy to implement.

This paper proposes to evaluate the installed performance of a slotted waveguide (SWA) antenna array on large platforms using an equivalent model based on near-field samplings over slots. The tangential electric fields over the SWA slots were first sampled to obtain equivalent magnetic current sources. The equivalent sources were then put on a closed perfect electric conducting box, same profile of the SWA, to form the equivalent model. This equivalent model avoids direct modeling of the complex inner structure and the thickness of the SWA. The installed radiation of the SWA on a platform was calculated by adding the direction radiation from the equivalent sources and the scattered fields from the induced currents over the whole antenna-on-platform system, where the induced currents were obtained by the method of moments. The effectiveness and efficiency was demonstrated via numerical cases.

A 10-bit, 100MS/s digital-to-analog converter (DAC) is designed in a standard 0.18μm CMOS process, where a segmented current-steering technique is employed. Through elaborative comparison among different decoding strategies, a “6+4” segmented architecture is used to obtain a balance between area cost and performance. In order for excellent performance, an array of current sources based on a regulated cascode current mirror technique is used to achieve both high output impedance and large-swing output voltage. A latch with functions of amplitude limiting and clock synchronization is employed to reduce glitches and improve dynamic performance. From a single 1.8 V supply, simulation results show that a full-scale current of 8 mA, a differential non-linear (DNL) of ±0.1LSB, an integral nonlinear error (INL) of ±0.4LSB, and a spurious-free dynamic range (SFDR) of 52dB are achieved.