Bearing Target Motion Analysis Method

Passive bearings-only target motion analysis (BO-TMA) is to estimate the position and velocity of a target from bearing measurements with corrupted noise obtained by an observation platform. Because of its simple algorithm and small amount of calculation, this method has been extensively used in many fields, such as navigation and aviation. In the thesis, based on the achievements have been acquired, the observability of two-dimensional BO-TMA system is discussed, and the location methods, algorithm and the estimation properties of target motion analysis with single array and multiple arrays are studied, and the simulation results are presented. All these work provide the theory foundation and evidence for the design of passive location system.Aiming at the problems appeared during the study process, the following work are carried out:1.Firstly, the mathmatical model of a target moved in a plane is established. The observability of the target is discussed using nonlinear system theory, and subsequently used to specify necessary and sufficient conditions on own-ship motion that insure a unique tracking solution.2.For special own-ship target maneuver geometry, target motion analysis is developed employing a maximum likelihood estimate (MLE), a pseudolinear estimate (PLE) and a modified instrumental variable (MIV) approach. A performance comparison is drawn between the PLE, the MIV and the MLE method.3.The application of Kalman filter in BO-TMA is studied. In the thesis, three Kalman filter methods are used, namely, the extended Kalman filter (EKF), the Pseudolinear (PL) filter and the modified polar (MP) coordinate filter. Different mathmatical models are established with varying estimation techniques. The corresponding filter equations are given and the estimation properties are analyzed theoretically.4.A new BO-TMA method combined adaptive weighting with Kalman filter is proposed. The algorithm is developed in detail by the use of Bayesian estimation technique. The block diagram of the algorithm structure is presented and its application in engineering is analyzed.5. Directing towards the bad observabihty of single array measurement location methods, double array TMA approaches are studied. First of all, a double-array data fusion model is presented, and the observability of the model is discussed. Under the condition that the target is observable, the MLE method and the Kalman filter approach with double-array measurements are studied.