An Integrated Radar-telemetry Fusion Method Based on Kalman Filtering

Traditional separate radar and telemetry systems suffer from equipment redundancy, insufficient synergy of multi-source heterogeneous data, and disconnection between tracking and data processing. Under the radar-telemetry integrated system architecture, an adaptive weighted fusion method using Kalman filtering is proposed in this paper. First, to address the dimensionality discrepancy among reflective position data, transponder position data, and telemetry angle-only data, linear and extended Kalman filters are respectively designed for local estimation, and an innovation χ² test is introduced for online outlier detection and removal. Second, the cross-covariance between the telemetry and the transponder local estimates is explicitly incorporated in the fusion center, and adaptive weighting factors are derived based on the minimum mean square error criterion, achieving a globally optimal track fusion. Finally, a high-precison system tracking is achieved by means of a closed-loop linkage architecture of "data fusion → tracking scheduling → servo control". Simulation results demonstrate that the proposed method maintains stable tracking performance in both normal and abnormal scenarios, with fusion accuracy significantly superior to that of any single sensor. The method effectively solves the problems of adaptive fusion of multi-source heterogeneous data and closed-loop linkage optimization, and offers strong engineering application value.