%创建跟踪场景。场景= trackingScenario (“StopTime”,30);%定义单位转换。nmi2m = 1852;%海里米的hr2 = 3600;%小时秒kts2mps = nmi2m /的hr2;%结米每秒

%创建监视雷达。传感器= fusionRadarSensor (1,“没有扫描”,…“MountingLocation”(0 0 -20),…% 20米(ASL)“MountingAngles”(45 0 0),…%(偏航距辊)度“FieldOfView”30 [10],…% (az el)度“ReferenceRange”5 e3,…% m“AzimuthResolution”2,…%度“RangeResolution”5,…% m“之内”,真的,…%报告INS信息“TargetReportFormat”,“检测”,…%检测没有集群“DetectionCoordinates”,“球形传感器”);

平台(场景中,“传感器”、传感器);塔= scenario.Platforms {1}

%定义的维度两个小的船只。昏暗的=结构(…“长度”,80,…% m“宽度”15岁的…% m“高度”5,…% m“OriginOffset”[0 0 5/2]);% (x y z) m%模型的雷达截面(RCS)的小型船只30 dBsm。rcs = rcsSignature (“模式”,30);%创建一个轨迹。速度= 20;%结initYaw = 130;%度initPos = (1050 790 0);半径= 200;% minitOrient =四元数([initYaw 0 0],“eulerd”,“ZYX股票”,“帧”);initVel =速度* kts2mps * rotatepoint (initOrient [1 0 0]) ';accBody =[0(速度* kts2mps) ^ 2 /半径0];angVelBody =[0 0速度* kts2mps /半径);traj = kinematicTrajectory (“位置”initPos,“速度”initVel,“定位”initOrient,…“AccelerationSource”,“属性”,“加速”accBody,…“AngularVelocitySource”,“属性”,“AngularVelocity”,angVelBody);%添加第一个小船,在20海里的场景。这是%雷达塔最近的船。平台(场景中,“维度”昏暗的,“签名”rcs,“轨迹”,traj);%创建其他小型船舶,30节的速度运行。这是这艘船从雷达塔%是最远的。速度= 30;%结initYaw = 120;%度initPos = (1410 1180 0);半径= 400;% minitOrient =四元数([initYaw 0 0],“eulerd”,“ZYX股票”,“帧”);initVel =速度* kts2mps * rotatepoint (initOrient [1 0 0]) ';accBody =[0(速度* kts2mps) ^ 2 /半径0];angVelBody =[0 0速度* kts2mps /半径);traj = kinematicTrajectory (“位置”initPos,“速度”initVel,“定位”initOrient,…“AccelerationSource”,“属性”,“加速”accBody,…“AngularVelocitySource”,“属性”,“AngularVelocity”,angVelBody);平台(场景中,“维度”昏暗的,“签名”rcs,“轨迹”,traj);%定义大型船舶的维度。昏暗的=结构(…“长度”,400,…% m“宽度”现年60岁的…% m“高度”15岁的…% m“OriginOffset”[0 0 15/2]);% (x y z) m%模型的雷达截面(RCS)的大型船舶75 dBsm。rcs = rcsSignature (“模式”,75);%创建大型船的轨迹,旅行以恒定10节。速度= 10;%结initYaw = -135;%度initPos = (1150 1100 0);initOrient =四元数([initYaw 0 0],“eulerd”,“ZYX股票”,“帧”);initVel =速度* kts2mps * rotatepoint (initOrient [1 0 0]) ';traj = kinematicTrajectory (“位置”initPos,“速度”initVel,“定位”initOrient,…“AccelerationSource”,“属性”,“AngularVelocitySource”,“属性”);%大型船舶添加到场景中。平台(场景中,“维度”昏暗的,“签名”rcs,“轨迹”,traj);%创建一个显示显示真实的测量和跟踪船只的位置。theaterDisplay = helperMarineSurveillanceDisplay(场景中,…“IsSea”,真的,“DistanceUnits”,“米”,…“XLim”,450 * [1]+ 1 e3,“YLim”,450 * [1]+ 1 e3,“ZLim”-1000年[10],…“电影”,“MarineSurveillanceExample.gif”);slctTrkPos = 0 (3、7);slctTrkPos (1, - 1) = 1;slctTrkPos (2、3) = 1;slctTrkPos (3、6) = 1;slctTrkVel = circshift (slctTrkPos [0, 1]);theaterDisplay。TrackPositionSelector = slctTrkPos;theaterDisplay。TrackVelocitySelector = slctTrkVel; theaterDisplay(); snapnow(theaterDisplay);

%定义雷达的测量范围和分辨率。azLimits = sensor.FieldOfView (1) / 2 * [1];%度rangeLimits = 15 e3 [0];% msensorLimits = [azLimits; rangeLimits];sensorResolution = [sensor.AzimuthResolution; sensor.RangeResolution];%定义传感器的安装位置和取向在塔上。参数(1)=结构(“帧”,“球”,…“OriginPosition”sensor.MountingLocation (:)…“OriginVelocity”(0,0,0),…“定位”rotmat(四元数(sensor.MountingAngles“eulerd”,“zyx股票”,“帧”),“帧”),…“IsParentToChild”,真的,…“HasRange”,真的,“HasElevation”sensor.HasElevation,“HasVelocity”、假);%定义塔的位置、速度和方向的场景。参数(2)=结构(“帧”,“矩形”,…“OriginPosition”tower.Trajectory.Position (:)…“OriginVelocity”tower.Trajectory.Velocity (:)…“定位”rotmat (tower.Trajectory.Orientation“帧”),…“IsParentToChild”,真的,…“HasRange”,真的,“HasElevation”假的,“HasVelocity”、假);%创建trackingSensorConfiguration模型的检测能力%由传感器跟踪。探测概率定义了概率%的生成至少1检测扩展对象。sensorConfig = trackingSensorConfiguration (“SensorIndex”sensor.SensorIndex,…“SensorLimits”sensorLimits,…“SensorResolution”sensorResolution,…“DetectionProbability”,0.99,…“SensorTransformParameters”、参数);%字段用于更新trackingSensorConfiguration传感器配置。configFlds = {“SensorIndex”,“IsValidTime”};%噪声协方差对应于一个分辨单元的雷达。resolutionNoise =诊断接头((sensorResolution / 2) ^ 2);sensorConfig。FilterInitializationFcn = @(变长度输入宗量)filterInitFcn(变长度输入宗量{:},params);sensorConfig。SensorTransformFcn = @ctmeas;sensorConfig。ClutterDensity = sensor.FalseAlarmRate / (sensor.AzimuthResolution * sensor.RangeResolution);%创建一个使用trackingSensorConfiguration博士跟踪器。追踪= trackerPHD (“SensorConfigurations”sensorConfig,…“HasSensorConfigurationsInput”,真的,…“PartitioningFcn”@ (x) partitionDetections (x, 1.5, 6),…“ExtractionThreshold”,0.75,…“DeletionThreshold”1 e-6…“出生率”1 e-5);

%初始化场景和追踪。重启(场景);重置(跟踪);%设置模拟雷达的发展更新的速度。场景。UpdateRate = sensor.UpdateRate;%设置随机种子可重复的结果。rng (2019“旋风”);%的模拟运行。snapTimes = 7 scenario.StopTime [2];%秒而推进(场景)%得到当前仿真时间。时间= scenario.SimulationTime;%从塔的生成检测雷达。[侦破,~,配置]=检测(塔、时间);%更新测量噪声的检测与雷达的决议。依据= updateMeasurementNoise(侦破,resolutionNoise);%更新追踪。trackSensorConfig = computeTrackingSensorConfig(配置、configFlds);跟踪=追踪(精细、trackSensorConfig、时间);%更新显示与当前梁位置、检测和跟踪的位置。theaterDisplay(引爆器,配置,跟踪);%的快照。snapFigure (theaterDisplay,任何(时间= = snapTimes));结束writeMovie (theaterDisplay);

showSnapshot (theaterDisplay, 1)

showSnapshot (theaterDisplay, 2)轴([1250 1450 1150 1350]);视图(90 [-90]);

showSnapshot (theaterDisplay, 3)轴([650 850 700 900]);视图(90 [-90]);

showSnapshot (theaterDisplay, 3)轴([900 1100 1250 1450]);视图(90 [-90]);

numTrks =元素个数(跟踪);TrackID = [tracks.TrackID] ';长度= 0 (numTrks, 1);宽度= 0 (numTrks, 1);身高= 0 (numTrks, 1);为iTrk = 1: numTrks ext =跟踪(iTrk) .Extent;[Q, D] = eig (ext);d = 2 *√诊断接头(d));iDims = 1:3;= (0 0 1);[~,iUp] = max (abs (* Q));高度(iTrk) = d (iDims (iUp));iDims (iUp) = [];长度(iTrk) = max (d (iDims));宽度(iTrk) = min (d (iDims));结束%显示表的估计尺寸的船只。dim =表(TrackID、长度、宽度、高度)

函数配置= computeTrackingSensorConfig (configIn, flds)配置=结构();为iFld = 1:元素个数教主thisFld = flds的{iFld};配置(thisFld) = configIn。(thisFld);结束结束

函数依据= updateMeasurementNoise(精细、噪音)为iDet = 1:元素个数(检波器)精细{iDet} .MeasurementNoise(:) =噪音(:);结束结束

函数博士= filterInitFcn (measParam变长度输入宗量)%这个函数使用预测出生密度只有模拟出生%的场景。如果输入参数个数= = 1%设置预测出生在视场均匀。phdDefault = initctggiwphd;% 1。使用方位和范围创建均匀分布状态。阿兹= 0;%所有组件在方位。范围= linspace (1000、10000、5);% 5的组件范围(阿兹,R) = meshgrid (Az,范围);%创建一个博士过滤器来分配内存。博士= ggiwphd(0(7元素个数(Az)), repmat(眼(7),[1 1元素个数(Az)]),…“ScaleMatrices”repmat(眼(3)[1 1元素个数(Az)]),…“StateTransitionFcn”@constturn,“StateTransitionJacobianFcn”@constturnjac,…“MeasurementFcn”@ctmeas,“MeasurementJacobianFcn”@ctmeasjac,…“PositionIndex”(1 3 6),“ExtentRotationFcn”phdDefault.ExtentRotationFcn,…“HasAdditiveProcessNoise”假的,“ProcessNoise”,2 *眼(4),…“TemporalDecay”1 e3,“GammaForgettingFactors”,1.1 *(1,元素个数(Az)),…“MaxNumComponents”,10000);为i = 1:元素个数(Az) [sensorX、感觉,sensorZ] = sph2cart(函数(阿兹(i)), 0, R (i));globalPos = measParam (1) .Orientation‘* [sensorX;感知;sensorZ] + measParam (1) .OriginPosition (:);博士学位。州(3 6][1,i) = globalPos;博士学位。StateCovariances (3 6 [1], [1 3 6] i) =诊断接头([1 e5 1000 e5]);使用位置协方差%覆盖组件之间的差距结束% 2。你所描述的“运动”国家的船只%的视野内。接下来,添加他们的大小和信息%的预期数量的检测。%%预计有两个类型的船只在海上,小和%。你每个尺寸创建组件。phdSmall =博士;%克隆博士过滤器对于大型船只。phdLarge =克隆(博士);%初始数量的组件集。numComps = phdSmall.NumComponents;%对小型船只,预计规模大约是100米长% 20米宽。取向是未知的,我们将创建4%为每个大小方向。首先,你必须将组件添加到%密度相同的状态。这可以通过简单地添加它小船%设置值追加(phdSmall phdSmall);追加(phdSmall phdSmall);%的自由度定义形状。大量代表%确定维度高。景深= 1000;%在vx协方差,v和ω。smallStateCov =诊断接头([300 300 50]);%矩阵规模小船smallShape =(景深- 4)*诊断接头([100/2计10]20/2。^ 2);% l, w和h%创建4方向45度。为我= 1:4 thisIndex =(张)* numComps + (1: numComps);R = rotmat(四元数((45 *(张)0 0),“eulerd”,“ZYX股票”,“帧”),“帧”);phdSmall.ScaleMatrices (:,:, thisIndex) = repmat (R * smallShape *”, [1 1 numComps]);phdSmall。StateCovariances ([2 4 5], [2 4 5], thisIndex) = repmat (R * smallStateCov *”, [1 1 numComps]);phdSmall。StateCovariances (7 [6]、[7] 6, thisIndex) = repmat(诊断接头(100 [100]),1 numComps [1]);结束%小型船只产生约10 - 20检测。expNumDets = 15;不确定性= 5 ^ 2;phdSmall.Rates (,) = expNumDets /不确定性;phdSmall.Shapes (,) = expNumDets ^ 2 /不确定性;phdSmall.DegreesOfFreedom(:) =景深;大型船舶%遵循类似的过程。追加(phdLarge phdLarge);追加(phdLarge phdLarge);largeStateCov =诊断接头((100 5));largeShape =(景深- 4)*诊断接头([500/2 100/2 10]。^ 2);为我= 1:4 thisIndex =(张)* numComps + (1: numComps);R = rotmat(四元数((45 *(张)0 0),“eulerd”,“ZYX股票”,“帧”),“帧”);phdLarge.ScaleMatrices (:,:, thisIndex) = repmat (R * largeShape *”, [1 1 numComps]);phdLarge。StateCovariances ([2 4 5], [2 4 5], thisIndex) = repmat (R * largeStateCov *”, [1 1 numComps]);phdLarge。StateCovariances (7 [6]、[7] 6, thisIndex) = repmat(诊断接头(100 [100]),1 numComps [1]);结束%生成大约100 - 200检测。expNumDets = 150;不确定性= 50 ^ 2;phdLarge.Rates (,) = expNumDets /不确定性;phdLarge.Shapes (,) = expNumDets ^ 2 /不确定性;phdLarge.DegreesOfFreedom(:) =景深;%添加大型船舶小型船只创建总密度。这个密度%添加到总密度每一步。博士= phdSmall;追加(博士,phdLarge);结束%称为时检测输入出生即自适应密度,不%添加任何新组件。如果输入参数个数> 1%这将创建0组件的密度。博士= initctggiwphd;结束结束