基于Riccati方程的动态传感器分配算法

摘要

A dynamic multi-sensor allocation algorithm based on the Riccati equation is proposed to deal with the computational burden of the sensor allocation problem in the multi-sensor tracking system. The algorithm calculates the stable covariance with the Riccati equation outline and dynamically allocates the optimal sensor sets based on the difference between the expected covariance and the stable covariance. Simulation results show that the algorithm can obtain the better tracking accuracy and computation efficiency comparing with the traditional covariance control algorithm and the greedy algorithm. The algorithm can be applied to the multi-sensor collaborative tracking in large-scale sensor networks.%针对组网跟踪系统传感器分配算法计算量过大的问题，提出了一种基于Riccati方程的动态传感器分配算法。该算法通过Riccati方程离线计算各传感器组合跟踪下的稳态滤波协方差，根据稳态滤波协方差与期望协方差的接近程度动态分配传感器资源。仿真结果显示，与传统协方差控制和贪婪算法相比，基于Riccati方程的动态传感器分配算法在大大减少计算量的同时能够保持较好的跟踪性能。该方法能够更好地应用于大规模传感器组网目标协同跟踪系统。