采用双目视觉和自适应Kalman滤波的作物行识别与跟踪

针对传统作物行识别方法在相邻图像间的识别结果偏差较大,作物行的定位精度和稳定性低等问题,该研究提出一种基于双目视觉和自适应Kalman滤波技术的作物行识别与跟踪方法。对于作物行识别,首先建立图像预处理算法,基于改进的超绿-超红模型和最大类间方差法分割植被灰度特征;建立作物行特征提取算法,基于特征点检测技术和双目视差测距方法计算植被角点特征的三维坐标,根据三维阈值提取作物行特征点,进而建立作物行中心线检测算法,建立基于主成分分析的直线拟合模型,根据作物行特征点的频数统计规律检测作物行冠层中心线。对于作物行跟踪,建立跟踪目标规划模型,提取位于图像中央区域的作物行作为跟踪目标;建立目标状态方程,基于自适应Kalman滤波技术构建作物行中心线跟踪模型。以棉花图像开展试验研究,图像数据包括阴影、杂草、地头等田间场景。试验结果表明,该研究方法的作物行识别准确度、精度和速度均较高,识别正确率约为92.36%,平均航向偏差为0.31°、标准差为2.55°,平均识别速度约80.25 ms/帧;经目标跟踪后,航向角和横向位置估计的标准差分别为2.62°和0.043m、较无跟踪状态分别减小22.94%和10.42%,作物行中心线的方位估计精度进一步提高。研究成果可为导航系统提供连续、稳定的作物行导引参数。

基于UKF的雷达和机载ESM扩维跟踪

为有效提高异类传感器融合跟踪的效果,提出了一种基于UKF的雷达和机载ESM扩维跟踪算法。首先将目标和机载ESM的状态信息组合起来,形成一个高维的状态向量,从而有效抑制滤波中GPS定位误差所带来的影响;接着在此基础上采用UKF来实现目标的定位跟踪,以求进一步减小扩维后线性化误差加大等问题的影响。仿真结果表明,该算法可更好地实现对目标的定位跟踪。

Video-based urban expressway traffic measurement and performance monitoring

This paper presents an urban expressway video surveillance and monitoring system for traffic flow measurement and abnormal performance detection. The proposed flow detection module collects traffic flow statistics in real time by leveraging multi-vehicle tracking information. Based on these online statistics, road operating situations can be easily obtained. Using spatiotemporal trajectories, vehicle motion paths are encoded by hidden Markov models. With path division and parameter matching, abnormal performances containing extra low or high speed driving, illegal stopping and turning are detected in real scenes. The traffic surveillance approach is implemented and evaluated on a DM642 DSP-based embedded platform. Experimental results demonstrate that the proposed system is feasible for the detection of vehicle speed, vehicle counts and road efficiency, and it is effective for the monitoring of the aforementioned anomalies with low computational costs.

Multi-object tracking based on behaviour and partial observation

To cope with multi-object tracking under real-world complex situations, a new video-based method is proposed. In the detecting step, the moving objects are segmented with the third level DWT （discrete wavelet transform ）and background difference. In the tracking step, the Kalman filter and scale parameter are used first to estimate the object position and bounding box. Then, the center-association-based projection ratio and region-association-based occlusion ratio are defined and combined to judge object behaviours. Finally, the tracking scheme and Kalman parameters are adaptively adjusted according to object behaviour. Under occlusion, partial observability is utilized to obtain the object measurements and optimum box dimensions. This method is robust in tracking mobile objects under such situations as occlusion, new appearing and stablization, etc. Experimental results show that the proposed method is efficient.

ORA_ROWSCN伪列在Oracle历史数据迁移中的应用研究

通过对Oracle常见数据迁移方法的利弊分析,给出利用ORA_ROWSCN行级跟踪模式的Oralce 10g数据迁移解决方案,该方案既使数据迁移省时省力,又能保证数据库的完整性和一致性,为企业做Oracle历史数据迁移提供完美方案。

ADAPTIVE NEURAL NETWORK ATTITUDE CONTROL FOR UNMANNED HELICOPTER

Adaptive flight control technology, feedback linearization, model inversion theory are reviewed and the error dynamic characteristics are analyzed, and an adaptive on-line neural network attitude control system is presented. The model inversion is under the hover condition. And the adaptive control law based on the neural network is designed to guarantee the boundedness of tracking error and control signals. Simulation results demonstrate that the nonlinear neural network augmented model inversion can self-adapt to the uncertainty and modeling errors of unmanned helicopters. Results are compared while the parameters of PD controller and robustness items are changed.

A robust system for real-time pedestrian detection and tracking

A real-time pedestrian detection and tracking system using a single video camera was developed to monitor pedestrians. This system contained six modules: video flow capture, pre-processing, movement detection, shadow removal, tracking, and object classification. The Gaussian mixture model was utilized to extract the moving object from an image sequence segmented by the mean-shift technique in the pre-processing module. Shadow removal was used to alleviate the negative impact of the shadow to the detected objects. A model-free method was adopted to identify pedestrians. The maximum and minimum integration methods were developed to integrate multiple cues into the mean-shift algorithm and the initial tracking iteration with the competent integrated probability distribution map for object tracking. A simple but effective algorithm was proposed to handle full occlusion cases. The system was tested using real traffic videos from different sites. The results of the test confirm that the system is reliable and has an overall accuracy of over 85%.

Three-dimensional neural network tracking control of autonomous underwater vehicles with input saturation

This paper addresses the problem of three-dimensional trajectory tracking control for underactuated autonomous underwater vehicles in the presence of parametric uncertainties,environmental disturbances and input saturation.First,a virtual guidance control strategy is established on the basis of tracking error kinematics,which resolves the overall control system into two cascade subsystems.Then,a first-order sliding mode differentiator is introduced in the derivation to avoid tedious analytic calculation,and a Gaussian error function-based continuous differentiable symmetric saturation model is explored to tackle the issue of input saturation.Combined with backstepping design techniques,the neural network control method and an adaptive control approach are used to estimate composite items of the unknown uncertainty and approximation errors.Meanwhile,Lyapunov-based stability analysis guarantees that control error signals of the closed-loop system are uniformly ultimately bounded.Finally,simulation studies are conducted for the trajectory tracking of a moving target and a spiral line to validate the effectiveness of the proposed controller.

Earthquake dynamic response behavior of Xiangchong valley type tailings impoundment in Yunnan, China

Tailings impoundments can potentially collapse due to damage caused by earthquakes,which has frequently occurred around the world.This study takes the proposed valley type tailings impoundment in Yunnan as the research object to analyze the dynamic response behavior under earthquake action with both numerical simulation and physical model test(1:300).The results of both tests show that the dynamic response of the valley type tailings impoundment is characterized by"medium stiffness effect",in other words,in a certain range,the"softer"the unsaturated tailings sand is,the more energy it can dissipate,which leads the decrease of the value of the acceleration amplification factor.In addition,the peak acceleration of the monitoring points increases with the vertical elevation,which indicates that the"elevation amplification effect"exists in the tailings impoundment dynamic response.The middle part of the outer side of the raised embankment reacts more sensitive than the crest,which is similar to the slope dynamic response.The starter dam reacts sensitively under the earthquake excitation,which should be given more attention during the seismic design.The dynamic response rules reflected by the numerical simulation are consistent with the results monitored on the physical model test,although there are some differences between their values.The dynamic response rules of the valley type tailings impoundment can provide basis for the design of the similar projects in this region.

Adaptive tracking control for a class of uncertain chaotic systems

The paper is concerned with adaptive tracking problem for a class of chaotic system with time-varying uncertainty, but bounded by norm polynomial. Based on adaptive technique, it proposes a novel controller to asymptotically track the arbitrary desired bounded trajectory. Simulation on the Rossler chaotic system is performed and the result verifies the effectiveness of the proposed method.

An LMI Method to Robust Iterative Learning Fault-tolerant Guaranteed Cost Control for Batch Processes

Based on an equivalent two-dimensional Fornasini-Marchsini model for a batch process in industry, a closed-loop robust iterative learning fault-tolerant guaranteed cost control scheme is proposed for batch processes with actuator failures. This paper introduces relevant concepts of the fault-tolerant guaranteed cost control and formulates the robust iterative learning reliable guaranteed cost controller (ILRGCC). A significant advantage is that the proposed ILRGCC design method can be used for on-line optimization against batch-to-batch process uncertainties to realize robust tracking of set-point trajectory in time and batch-to-batch sequences. For the convenience of implementation, only measured output errors of current and previous cycles are used to design a synthetic controller for iterative learning control, consisting of dynamic output feedback plus feed-forward control. The proposed controller can not only guarantee the closed-loop convergency along time and cycle sequences but also satisfy the H∞performance level and a cost function with upper bounds for all admissible uncertainties and any actuator failures. Sufficient conditions for the controller solution are derived in terms of linear matrix inequalities (LMIs), and design procedures, which formulate a convex optimization problem with LMI constraints, are presented. An example of injection molding is given to illustrate the effectiveness and advantages of the ILRGCC design approach.

An efficient human detection method for multi-pedestrian tracking

Traditional human detection using pre-trained detectors tends to be computationally intensive for time-critical tracking tasks, and the detection rate is prone to be unsatisfying when occlusion, motion blur and body deformation occur frequently. A spatial-confidential proposal filtering method(SCPF) is proposed for efficient and accurate human detection. It consists of two filtering phases: spatial proposal filtering and confidential proposal filtering. A compact spatial proposal is generated in the first phase to minimize the search space to reduce the computation cost. The human detector only estimates the confidence scores of the candidate search regions accepted by the spatial proposal instead of global scanning. At the second phase, each candidate search region is assigned with a supplementary confidence score according to their reliability estimated by the confidential proposal to reduce missing detections. The performance of the SCPF method is verified by extensive tests on several video sequences from available public datasets. Both quantitatively and qualitatively experimental results indicate that the proposed method can highly improve the efficiency and the accuracy of human detection.

Chaos Existence in Surface Discharge of Tracking Test

Tracking tests for different polymer materials were carried out to investigate the chaotic behavior of surface discharge. The discharge sequences measured during the discharge process were analyzed for finding the evidence of chaos existence. Four kinds of nonlinear analysis methods were adopted： estimating the largest Lyapunov exponent, calculating the fractal dimension with increasing the embedding dimension, drawing the recurrence plots, and plotting the Poincare maps. It is found that the largest Lyapunov exponent of the discharge is positive, and the plot of fractal dimension, as a function of embedding dimension, will saturate at a value. The recur- rence plots show the chaotic frame-work patterns, and the Poincar6 maps also have the chaotic characteristics. The results indicate that the chaotic behavior does exist in the discharge currents of the tracking test.

Hybrid tracking model and GSLM based neural network for crowd behavior recognition

Crowd behaviors analysis is the‘state of art’research topic in the field of computer vision which provides applications in video surveillance to crowd safety,event detection,security,etc.Literature presents some of the works related to crowd behavior detection and analysis.In crowd behavior detection,varying density of crowds and motion patterns appears to be complex occlusions for the researchers.This work presents a novel crowd behavior detection system to improve these restrictions.The proposed crowd behavior detection system is developed using hybrid tracking model and integrated features enabled neural network.The object movement and activity in the proposed crowded behavior detection system is assessed using proposed GSLM-based neural network.GSLM based neural network is developed by integrating the gravitational search algorithm with LM algorithm of the neural network to increase the learning process of the network.The performance of the proposed crowd behavior detection system is validated over five different videos and analyzed using accuracy.The experimentation results in the crowd behavior detection with a maximum accuracy of 93%which proves the efficacy of the proposed system in video surveillance with security concerns.

Nonlinear trajectory tracking control of a new autonomous underwater vehicle in complex sea conditions

Autonomous underwater vehicles (AUVs) navigating in complex sea conditions usually require a strong control system to keep the fastness and stability. The nonlinear trajectory tracking control system of a new AUV in complex sea conditions was presented. According to the theory of submarines,the six-DOF kinematic and dynamic models were decomposed into two mutually non-coupled vertical and horizontal plane subsystems. Then,different sliding mode control algorithms were used to study the trajectory tracking control. Because the yaw angle and yaw angle rate rather than the displacement of the new AUV can be measured directly on the horizontal plane,the sliding mode control algorithm combining cross track error method and line of sight method was used to fulfill its high-precision trajectory tracking control in the complex sea conditions. As the vertical displacement of the new AUV can be measured,in order to achieve the tracking of time-varying depth signal,a stable sliding mode controller was designed based on the single-input multi-state system,which took into account the characteristic of the hydroplane and the amplitude and rate constraints of the hydroplane angle. Moreover,the application of dynamic boundary layer can improve the robustness and control accuracy of the system. The computational results show that the designed sliding mode control systems of the horizontal and vertical planes can ensure the trajectory tracking performance and accuracy of the new AUV in complex sea conditions. The impacts of currents and waves on the sliding mode controller of the new AUV were analyzed qualitatively and quantitatively by comparing the trajectory tracking performance of the new AUV in different sea conditions,which provides an effective theoretical guidance and technical support for the control system design of the new AUV in real complex environment.

Parallel Open Source CFD-DEM for Resolved Particle-Fluid Interaction

A parallelized resolved method for the simulation of the dynamics of immersed bodies within fluids is presented. The algorithm uses a FDM （fictitious domain method） and combines the Lagrangian DEM （discrete element method） for tracking the bodies with a CFD （computational fluid dynamics） method for calculating the dynamics of the fluid phase. First the CFD-calculation is carried out, disregarding the solid bodies. Afterwards, the velocity information from the bodies is included and the force, the fluid imposes onto the bodies, is computed. The last step consists of a correction-operation which ensures the fulfillment of the conservation equation. Dynamic local mesh refinement is used for minimizing the number of fluid cells. The CFD-DEM coupling is realized within the Open Source framework CFDEMcoupling （www.cfdem.com）, where the DEM software LIGGGHTS （www.liggghts.com） is linked against an OpenFOAM^-based CFD solver. While both LIGGGHTS and the CFD solver were already parallelized, only a recent improvement of the algorithm permits the fully parallel computation of resolved problems. This parallelization permits the treatment of large-scale problems. The enclosed validation and application examples show the dynamics of the flow around settling and rotating spheres as well as an investigation of the settling of spheres regarding the Boycott effect.

Following Control for a UUV using Temporary Path Generation Guidance

A path following control algorithm for an unmanned underwater vehicle（UUV） using temporary path generation guidance was proposed in this paper.Owing to different initial states of the vehicle,such as position and orientation,the path following control in the horizontal plane may yield a poor performance.To deal with the negative effect induced by initial states,a temporary path generation was presented based on the relationship between the original reference path and the vehicle’s initial states.With different relative positions between the vehicle and reference path,including out of straight lines,as well as inside and outside a circle,the related temporary paths guiding the vehicle to the reference path were able to be generated in real time.The vehicle was guided to steer along the temporary path until it reached the tangent point at the reference path,where the controller was designed using the input-output feedback linearization method.Simulation results demonstrated that the proposed algorithm is effective under the three different situations mentioned above.

Robust tracking control design for a flexible air-breathing hypersonic vehicle

A nonlinear robust controller was presented to improve the tracking control performance of a flexible air-breathing hypersonic vehicle(AHV) which is subjected to system parametric uncertainties and unknown additive time-varying disturbances.The longitudinal dynamic model for the flexible AHV was used for the control development.High-gain observers were designed to compensate for the system uncertainties and additive disturbances.Small gain theorem and Lyapunov based stability analysis were utilized to prove the stability of the closed loop system.Locally uniformly ultimately bounded tracking of the vehicle's velocity,altitude and attack angle were achieved under aeroelastic effects,system parametric uncertainties and unknown additive disturbances.Matlab/Simulink simulation results were provided to validate the robustness of the proposed control design.The simulation results demonstrate that the tracking errors stay in a small region around zero.

Road boundary estimation to improve vehicle detection and tracking in UAV video

Video processing is one challenge in collecting vehicle trajectories from unmanned aerial vehicle(UAV) and road boundary estimation is one way to improve the video processing algorithms. However, current methods do not work well for low volume road, which is not well-marked and with noises such as vehicle tracks. A fusion-based method termed Dempster-Shafer-based road detection(DSRD) is proposed to address this issue. This method detects road boundary by combining multiple information sources using Dempster-Shafer theory(DST). In order to test the performance of the proposed method, two field experiments were conducted, one of which was on a highway partially covered by snow and another was on a dense traffic highway. The results show that DSRD is robust and accurate, whose detection rates are 100% and 99.8% compared with manual detection results. Then, DSRD is adopted to improve UAV video processing algorithm, and the vehicle detection and tracking rate are improved by 2.7% and 5.5%,respectively. Also, the computation time has decreased by 5% and 8.3% for two experiments, respectively.