Behavioural transition during the estuarine migration of wild Atlantic salmon (Salmo salar L.) smolt

Abstract –  Ultrasonic telemetry and hydrodynamic modelling were used to study the migratory behaviour of 54 wild Atlantic salmon ( Salmo salar ) smolt captured in freshwater during their downstream migration and tracked in 2 years through a shallow estuary system. A high-density, fixed array of receivers provided detailed spatial and temporal resolution of behaviour in the second year of study. Smolt migration in the river occurred mostly at night and downstream migration was slower during the day. In the estuary, smolt moved seaward on ebbing tides and landward on flooding tides. The effect of current velocity was greater during the night than during the day. We documented for the first time that current velocity and diurnal period only accounted for approximately one-third of the variation in smolt ground speeds in the estuary, indicating that smolt movements were far less passive than previously reported. Smolt energetic status had no effect on smolt swimming behaviour or migratory performance. With an increase in salinity, smolt seaward movements during flooding tides were more frequent, and overall seaward ground velocity increased. The increase in salinity experienced by the smolt during their migration through the leading edge of saltwater intrusion thus induced a behavioural transition from a more passive, fluvial migration to a more active- and seaward-oriented migration.     

低频超声二次雾化喷头智能驱动电路开发

低频超声二次雾化喷头通过声悬浮实现二次雾化,其工作负载较为复杂,常规超声驱动电路难以连续可靠地驱动其工作。应用EVC(Embedded Visual C++),在基于ARM9.0的WINCE 5.0平台下开发了该喷头的智能驱动程序,实现了频率和占空比连续调节,并设计了喷头谐振频率自动跟踪电路。根据跟踪结果,改变ARM的PWM输出频率,可保证喷头工作在谐振点附近;通过对该智能驱动电路的Simulink仿真,验证了其正确性。     

拖拉机沿曲线路径的跟踪控制

近年关于农业用车轮型移动机器人的研究很多,但主要集中在直线行走方面。该研究进行了沿给定的曲线路径跟踪控制的研究。首先由给定的曲线路径生成四元状态空间,其次在利用预见控制求得车辆的未来值和目标值的基础上,利用最优控制理论设计跟踪控制器。最后在牧草地上对正弦路径和圆形路径进行了跟踪实验。试验结果为：车辆以1.5 m/s的速度行驶时,在很小转弯半径时最大误差可控制在35 cm以内。     

基于卡尔曼滤波器的运动目标检测与跟踪

针对摄像机静止的情况,提出了一种可运用于实时监控中的运动目标检测与跟踪的方法．采用更新函数实现背景实时更新,通过差分算法检测运动目标．在跟踪模块中,提出建立帧间目标“关系矩阵”实现多个运动目标匹配,并采用卡尔曼滤波器预测目标参数,在运动目标相互遮挡的情况下,根据预测参数跟踪目标,获得目标轨迹．通过多个图像序列测试,算法具有良好的实时性和适应环境变化的能力．     

Machine Vision Aided Testing System for Flatness Based on Equal Inclination Interference Theory

An approach is described on the automated tracking of interference circle fringe to get the flatness accurately based on the Equal Inclination Interference Theory (EIIT). The system,based on the theory of EIIT, can be applied to grasp the changing fringe information, which is white alternating with black through the CCD camera. Following the basic principle of fringes image centro-symmetric, we create some windows, process only the interference fringes information in the windows. According to the position of the processed interference fringe and the position of the last adjacent interference fringe in the windows, the interference fringe variation value N from the starting measured point is received, it implies the changed altitude value from the starting measurement point. This method allows us not only get the fringe information of the half-wavelength integer multiple but also denote those of the half-wavelength decimal multiple. The system is proved to be effective experimentally. It is useful in metrology department for the high-precision flatness standard unit.     

Research of Holospectrum Technique Based on Order Tracking

This paper analyzes the principle of THS(Traditional Holo-Spectrum) technique,and in allusion to the deficiency of THS that only can be used in stationary signal,a new holo-spectrum technique based on order tracking is proposed.By order tracking sampling technology,non-stationary signal in the time domain is transformed into stationary signal in the angle domain,then the THS technique can be applied.Through this way,the influence of non-stationary to THS is reduced and the application of THS is extended.The paper researches the principle of order tracking holo-spectrum and verified by a simulation experiment.The simulation result and error analysis demonstrate that this method can be valid used to analyze the non-stationary signal.The spectrum is distinct and the effect is apparent.     

池塘养殖船自动导航系统设计与试验

针对池塘养殖中投喂作业需要全塘均匀覆盖的应用场景,存在人工投饲强度大、饲料利用率低的问题,设计一种能够适应不同多边形池塘的养殖船自动导航控制系统。控制系统采用低成本北斗定位模块和高精度电子罗盘进行组合导航,获取池塘养殖船的位置和航向信息作为导航控制器的输入,通过构建基于PD算法的导航控制器,实现航行过程中的路径跟踪。设计一种多边形回纹线导航路径规划算法,能够快速实现多边形池塘的导航路径规划。开展池塘导航试验,试验结果表明：采用所设计的自动导航系统,养殖船能够按照规划的路径航行,在水面行驶速度为0.4~0.5 m/s时,稳定跟踪后最大误差小于2.62 m,平均跟踪误差小于1.30 m,导航精度满足池塘养殖自动投饲要求。     

GB/T19630标准有机柑橘生产质量安全管理及追踪体系建立研究

对如何建立有机柑橘的安全管理及可追溯系统进行探讨。当有机柑橘生产与加工过程中出现操作违反准则的情况时,可以在最短的时间内,以最快的速度查找违规事项的原因,确认因违反准则需要回收的货物,并使其回收量降到最少,从而减少损失,确保有机柑橘的质量安全。     

基于SUKF与SIFT特征的红外目标跟踪算法研究

针对复杂红外背景下单一跟踪算法难以准确定位运动目标的问题,提出了基于尺度无迹卡尔曼滤波(SUKF,scale unscented Kalman filter)与尺度不变特征变换(SIFT,scale invariant featuretransform)相结合的红外运动目标跟踪方法。首先,通过SUKF算法对状态空间进行滤波估计,确定运动目标的初步位置,并以此建立局部SIFT特征检测域。其次,SIFT算法在该局部检测域内对运动目标进行特征提取与匹配,最终实现对目标的准确定位;同时,利用定位结果更新并校正SUKF的状态模型。实验结果表明,本文提出的基于SUKF-SIFT的跟踪策略与相关算法相比,体现出较好的跟踪效果与实时性能。