Abstract: To address the issues of imbalance between resource supply and demand and the poor adaptability of single management strategies in multi-radar systems, this paper proposes a covariance-information entropy hybrid closed-loop fusion strategy. By constructing a load-factor-based switching closed loop, a dynamic switching criterion based on the system load factor is designed: covariance control is adopted to ensure tracking accuracy when resources are sufficient, while information entropy control is switched to for improving real-time performance when resources are constrained. Monte Carlo simulations demonstrate that, in resource-critical scenarios, the proposed strategy reduces the position RMSE by 16.4% compared to the single covariance-based strategy and by 6.1% compared to the information entropy-based strategy, effectively balancing accuracy, resource utilization, and computational time. This work provides an engineering reference for collaborative multi-radar detection.