Tracking of Maneuvering Non-Ellipsoidal Extended Object or Target Group Using Random Matrix

Abstract

For non-ellipsoidal extended object and group target tracking (NEOT and NGTT), using a random matrix to simplify the extension as an ellipsoid, although efficient, may not be accurate enough because of the loss of useful information in shape and orientation. In view of this, we model a non-ellipsoidal extended object or target group as a combination of multiple ellipsoidal sub-objects, each represented by a random matrix. Different sub-objects are distinguished by different dynamic models. Based on such models, a Bayesian approach is proposed to estimate the kinematic states and the extensions of the sub-objects. For maneuvering NEOT and NGTT, a multiple-model approach is proposed. The overall maneuver model is combined with the sub-object models to describe the common and the individual motions of the sub-objects. Three simplifications of the approach are presented. Two scenarios for maneuvering and nonmaneuvering NEOT and NGTT are simulated. The results demonstrate the effectiveness of the proposed modeling and estimation approach.Index Terms-Bayesian approach, extended object, group target, multiple-model approach, random matrix.

Tracking of non-ellipsoidal extended target using negative measurements

“…By comparing RMSE of the target states, the proposed method is superior to methods in [7] and [8], as shown in Fig.1.…”

“…The scenario and parameters in [7] are used, where a plane-like target moves with the speed 260 m/s and the sampling period is 1s. The random matrix of the ellipse corresponding to the fuselage is 2 2 diag( [10 ,2 ]) , and the matrix of the wings is 2 2 diag( [10 ,5 ]) .…”

“…Measurement rates of the fuselage and the wings are 40 and 20, respectively. The root mean square errors (RMSE) of Lan's methods in [7] (denoted M1), Karl's method in [8] (denoted M2) and the proposed method (denoted M3) are shown in Fig. 1.…”

“…The number of measurements generated by the target is assumed to be Poisson distributed. Let Applying the non-ellipsoidal model in [7], an extended target is expressed by e N ellipsoidal sub-objects. Assuming a rigid body, these sub-objects share some common dynamics, which constrain the dynamics of each one.…”

An extended target particle probability hypothesis density filter based on the Star-Convex shape estimation

“…We assume the extended targets with Star-Convex shape, the observation area x~ [- Simulation experiment scenario 1 In order to fully verify the initiation performance of the proposed algorithm, the experiment compare the proposed initiation algorithm with the algorithm proposed by reference [9], as shown in Table 1, four moving targets appear one by one, the motion parameters set as follows. Fig.3 The track initiation diagram of extended target centroid based on 2/3 logic Fig.…”