Aiming to improve radar multi-target tracking (MTT) accuracy and association performance in complex scenarios involving dense clutter, missed detections, and maneuvering targets, an improved tracklet generation approach based on the expectation–maximization (EM) framework is proposed in which data association variables and motion model variables are jointly modeled as latent variables. These variables are estimated through iterative updates based on the loopy belief propagation (LBP) algorithm and the interacting multiple model (IMM) filtering and smoothing algorithms to generate high-confidence tracklets. Then, a delayed decision-making strategy based on the multi-hypothesis approach is employed to associate these tracklets into complete target trajectories. The resulting algorithm is named IMM-TrackletMHT. The performance of the IMM-TrackletMHT algorithm is evaluated and compared with several baseline algorithms in simulated scenarios under different clutter rates and detection probabilities. The simulation results demonstrate that the proposed algorithm consistently outperforms the baseline methods in terms of tracking accuracy, exhibits strong robustness to variations in the operating environment, and achieves higher computational efficiency in multi-scan measurement processing, thereby demonstrating the effectiveness and superiority of the proposed tracklet generation approach for maneuvering MTT.
Extended object tracking, which can obtain rich information about the object’s shape, size and orientation, has received widespread attention. The modeling method based on support functions or extended Gaussian image needs the down-range measurement and cross-range measurement of the object to estimate the information about the object’s extent. Down-range and cross-range refer to the projection lengths of the object along and perpendicular to the radar line of sight, respectively. The down-range measurement can be generated from the high range resolution profile. However, in some radar applications, limited by the resolution capability in cross-range, the cross-range measurements are unavailable. To track extended objects only using down-range measurements, this paper analyzes the observability of the extended object in detail. The analysis results indicate that the observability of the extended object is affected by the dimensions of range extent measurement, and the required dimensions of the range extent measurement for extended object tracking vary under different motion modes. According to the analysis results, when the object performs coordinate turn motion, extended object tracking can be achieved using only down-range measurements. The results of simulation experiments confirm this conclusion. The analysis results presented in this paper clarify the measurement conditions required for extended object tracking and expand the application scope of tracking algorithms based on the support function and extended Gaussian image modeling.