Research On Target Localization And Single Sensor Trajectory Planning With Constraints Based On Bearing-Only Measurements

With the popularization of mobile terminals,various locationbased services need to determine accurate positioning information,such as public safety services,field rescue and other fields.Bearing-only localization refers to the estimation of target position by collecting the angle measurements between the sensor and the target.This thesis studies the basic theory of bearing-only localization and trajectory planning.A bias compensation method and two sensor trajectory planning strategies are proposed to improve the positioning accuracy.Aiming at the problem of low positioning accuracy caused by the existence of measurement noise in bearing-only positioning system,this thesis presents a bias compensation method based on pseudo-linear estimation method.Under the condition of Gaussian noise,a potential relationship exists between the determinant of the extended coefficient matrix and the noise variance in the angle measurement equation.When the determinant of the extended coefficient matrix is zero,the variance of the zero-mean Gaussian noise can be derived.Based on this estimated variance,the bias compensation localization of the target on the twodimensional plane is realized.Then the bias compensation localization of the target in the z-axis direction is obtained analogously to the twodimensional positioning result.It is verified by simulation that the bias compensation method has lower the mean square error than the pseudolinear estimation method.Aiming at the problem of high computational complexity in optimizing the determinant of Fisher information matrix under the whole time span in trajectory planning,in this thesis,according to the property that the determinant of Fisher information matrix increments within a time step,two trajectory planning methods based on greedy strategy are proposed to realize the trajectory planning of a single sensor in twodimensional plane and three-dimensional space,respectively.Constraints such as the target safety area and the sensor movement area are considered in the planning.The optimal position of sensor at the next moment is optimized by using determinant increment as the objective function.In the first trajectory planning method,the position estimation of the target at the current moment is substituted into the objective function to update the angle information at all past moments.In the second trajectory planning method,the position estimation of the target at the current moment is only used to obtain the information matrix of the current moment,without updating the angle information at all past moments.Simulation results show that the two trajectory planning methods proposed in this thesis can obtain more information than other methods,and more accurate location results.Compared with the first method,the second method uses less precise information in exchange of faster calculation speed.