In the Kalman filter (KF), the estimated state is a linear combination of the one-step prediction and measurement. The two combination weights depend on the prediction mean-square error matrix and the covariances matrix of measurement noise (CMMN), respectively. When the measurement noise values are small (large), the corresponding measurement is close to (far from) the real value, and the weight should be large (small). If there is non-Gaussian measurement noise, especially heavy-tailed noise, most of the noise values are small, and a few of them are large. The occasional impulses cause the overall noise variance to be much greater than the noise without impulses. Since the overall CMMN is adopted in the KF, its performance will deteriorate. In this article, we divide the measurement noise into two parts: 1) the background noise with small variance and 2) the impulse noise with large variance. Then, we use the expectation–maximization algorithm to dynamically calculate the parameters and propose a new KF algorithm to process them separately, which is called background-impulse KF (BIKF). It can dynamically determine whether there is an impulse and eliminate its impact as much as possible. Compared with the recent maximum correntropy criterion and minimum error entropy criterion-based KFs, simulations show that the proposed BIKF works better than them with lower computational complexity.

Traditional interacting multiple model Kalman filter (IMM-KF) was derived for linear discrete time system with Markovian coefficients and it works well under Gaussian noise. In practice, when the system is affected by the impulsive noise, the performance of the traditional IMM-KF model becomes poor. Then, maximum correntropy Kalman filter (MCKF) which has advantages in dealing with non-Gaussian noise was proposed. However, the performance of traditional MCKF becomes poor when the motion state is not single. In this paper, we propose a new algorithm called interacting multiple model based on maximum correntropy Kalman filter (IMM-MCKF) by combining interacting multiple model (IMM) with MCKF to deal with the impulsive noise. Also, the traditional MCKF works poorly when the kernel bandwidth is small, we made a further improvement by changing the kernel. Simulation examples are given to demonstrate the effectiveness of the new algorithms.

In the data association of multi-target tracking, the probabilistic multi-hypothesis tracker (PMHT) algorithm has attracted the attention of many scholars because of its strong extensibility and low computational complexity. The traditional PMHT with kalman filter performs well under the Gaussian condition. However, in practical applications, the impulsive noise environment is common, in which case the performance of the PMHT becomes poor. In this paper, we apply the maximum correntropy criterion (MCC) and minimum error entropy (MEE) criterion to PMHT respectively to deal with the non-Gaussian noise. Simulation results show the effective performance of the new algorithms under multi-target and multi-sensor conditions.

Low frame-rate image sequences, as an emerging visual information carrier, are favored for mobile platforms due to their low redundancy, storage, and transmission requirements. However, their poor spatial-temporal coherence poses challenges for visual tracking tasks, affecting the accuracy of passive target localization for unmanned aerial vehicles. This paper suggests a joint registration and tracking module to tackle this problem. The module incorporates a SIFT-based image registration submodule to enhance the accuracy of target prediction, thus improving the robustness of tracking and the localization accuracy of passive target localization technology. Experimental results demonstrate the method’s superior robustness in target tracking, achieving a target localization accuracy of approximately 4.5 meters at a flying height of 100 meters. Additionally, the data storage requirements are reduced by a factor of 29.3, significantly decreasing onboard computing and transmission resource demands. This approach provides crucial support for enhancing the accuracy and robustness of unmanned aerial vehicle target localization and is expected to yield substantial benefits in practical applications, playing a pivotal role in enhancing the capabilities and effectiveness of unmanned aerial vehicle missions.