Multi-Sensor Fusion Tracking for Low-Slow-Small Targets Using Square-Root Cubature Kalman Filter

Low, Slow, and Small (LSS) targets, characterized by low altitude, slow speed, and small size, pose significant challenges for detection and tracking due to strong concealment and complex environmental interferences like multipath effects and electromagnetic noise. Current solutions focus on improving filtering algorithms and multi-sensor fusion, but face issues such as linearization errors, high computation, particle degeneration, and limited adaptability to anomaly detection under multipath interference. This paper addresses accurate position and velocity estimation of LSS targets by proposing a multi-sensor fusion tracking method within the SRCKF framework. It investigates three strategies—Covariance Intersection (CI) fusion, sequential fusion, and optimal subset fusion—leveraging SRCKF’s numerical stability and efficiency. Key contributions include a weighted statistical model for LSS motion, a multipath anomaly detection mechanism via innovation consistency testing, and an integrated data processing framework combining these fusion strategies to enhance tracking robustness and accuracy in complex environments.