Search

This study proposes an intelligent tracking method based on the expectation maximisation approach. To obtain the consistent tracking performance, the observer should maintain contact with the target and continuously update its tracking data. Unfortunately, a significant problem arises from the uncertainty of the target motion. The reasons behind this uncertainty include losses, time-varying noises, external inputs, and others. Above all, a sharp manoeuvre by the arbitrary acceleration input mainly degrades the tracking performance. A variety of techniques were needed and studied to enhance performance levels. This study focuses on acceleration as the cause of the uncertainty and a consistent approximation of the acceleration, leading to highly efficient tracking performance. The EM makes it possible to separate the noise term from the acceleration inputs and the noise components. Finally, a manoeuvring submarine with three-dimensional coordinates is provided as an example to show the effectiveness of the proposed algorithm.

The effort to detect a target conflicts with the covertness in anti-submarine warfare. The underwater detection rate is extremely low and sound waves are the only detection method unlike the airborne operational environment. The data of an underwater target based on noise levels are more difficult to interpret when the acceleration inputs are included. Hence, the authors propose a complementary algorithm for the tracking problem in an active sonar system. The bearing-range data of the target yield the measurement residuals, and the residuals are used as inputs to the clustering method. The clustering method estimates the acceleration and the noise level. The estimates aid in positional compensation and enhance the tracking performance. The fuzzy c-means clustering is utilised for the clustering and maintains the filter sustainable. Finally, some examples are provided to show the effectiveness of the proposed method.

This study proposes a smart tracking model with multiple structures. To achieve optimal performance, an ideal tracking system should maintain contact with the target and constantly update its tracking data. One of the major problems to consider arises from the target motion uncertainty. Although the variance of the overall noise is time-varying and may reduce tracking efficiency, it is difficult to adaptively approximate this uncertainty. To solve these computational problems, a variety of techniques have been studied, and the adaptive interacting multiple model has shown the best result. However, predefined sub-models with different conditions for multiple models are required. Moreover, the acceleration intervals for different acceleration levels should be determined in advance. To alleviate these problems, the authors propose a multiple-structured smart tracking algorithm which properly separates the acceleration from the overall noise by fuzzy c-means clustering. The filter in the algorithm recognises the manoeuvring target as if it does not include the acceleration input, and precisely tracks the target with multiple structures. This study focuses on the acceleration causing the uncertainty and three-dimensional (3D) motion of the real target motion in the tracking problem. Finally, a 3D cruise missile example is provided to show the effectiveness of the proposed algorithm.

This paper presents the smart interacting multiple model (SIMM) using the concept of predicted point and maximum noise level. Maximum noise level means the largest value of the mere noises. We utilize the positional difference between measured point and predicted point as acceleration. Comparing this acceleration with the maximum noise level, we extract the acceleration to recognize the characteristics of the target. To estimate the acceleration, we propose an optional algorithm utilizing the proposed method and the Kalman filter (KF) selectively. Also, for increasing the effect of estimation, the weight given at each sub-filter of the interacting multiple model (IMM) structure is varying according to the rate of noise scale. All the procedures of the proposed algorithm can be implemented by an on-line system. Finally, an example is provided to show the effectiveness of the proposed algorithm.