رديابي اهداف هوايي در صورت گم شدن اطلاعات فاصله

چكيده انگليسي :

Aerial Target Tracking While Missing Range Hamed Rabiee hamed3156@gmail com March 15 2011 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Persian Saeed Hosseinia hoseinia@cc iut ac ir Farid Sheikholeslam sheikh@cc iut ac irAbstract Target Tracking is an important problem in defense systems It means to obtain position and velocityof the target using noise corrupted measurements Target tracking can be divided into various categories These categories differ from each other in the type of target e g ship airplane missile and etc and type ofmeasured data for example the problem of tracking the target whose angle is only measured is calledbearing only tracking In this thesis we would talk about targets which are controlled by human or targets thattheir acceleration is smaller than 5g 7g but targets such as missiles aren t considered We would investigatevarious methods for solving major challenges existing in the problem of target tracking These challengesare 1 Target motion uncertainty in the problem of target tracking target trajectory dynamic model isunknown to the tracker so the tracker must consider some assumptions for the type of motion or trajectory ofthe target Usually trackers consider two or three types of motion for target These trajectories are A Constant Velocity CV in this type of motion tracker assumes that target follows a straight line with aconstant velocity B Constant Acceleration CA in this type of motion tracker assumes that target has aconstant acceleration whether following a straight line or not C Coordinate Turn or Constant Turn CT inthis type of motion tracker assumes that target is turning with a constant speed and constant turn rate constant angle rate 2 Measurement model uncertainty usually in the problem of target tracking it is betterto describe the dynamic model of target in Cartesian coordinate system But there is a problem that is thetracker can only measure position of the target in spherical coordinate system that causes a nonlinear relationbetween measurement and state vectors Position of the target is usually measured by radar or electro opticalsystems In electro optical systems angles are measured by two encoders and range is measured by a laserrangefinder The challenges mentioned above always occur in most of radar systems But in electro opticalsystems two other challenges in addition to above challenges occur These challenges are 3 In electro optical systems laser rangefinder measures the range of target with a different period from the period ofencoders For example in a typical electro optical system laser rangefinder measures the range of target withsampling time of 480 ms but encoders measure the angles of target with sampling time of 40 ms 4 Inelectro optical systems range of target may be missed Usually missing range of target occurs when targetstarts maneuver Because when the target starts maneuver the beam of rangefinder doesn t always hit it sothe rangefinder can t measure range of it In this thesis some solutions for the mentioned challenges arestated For solving the first challenge we could use Multiple Model Estimators In these estimators multiplemodels are considered for the system that its state should be estimated These estimators are divided intovarious categories We use Interaction Multiple Model Estimator IMM for solving the first challenge There

ربيعي، حامد

رديابي اهداف هوايي در صورت گم شدن اطلاعات فاصله

تعقيب هدف , تخمين گر چند گانه , تبديل بدون باياس , BLUE فيلتر