He Liu, Zhao Yingjun, Dong Qingkuan
Air and Missile Defense College, Air Force Engineering University, Xi'an 710051, China.
State Key Laboratory of Integrated Services Networks (ISN), Xidian University, Xi'an 710071, China.
Heliyon. 2020 May 16;6(5):e03832. doi: 10.1016/j.heliyon.2020.e03832. eCollection 2020 May.
In this study, an event-driven state estimator is designed for stochastic systems that contain unknown inputs and processes as well as correlated measurement noise. First, the event-triggered state estimator's gain is deduced by using the random stability theory and Lyapunov's function. Then, based on the results, the corresponding state estimation errors are calculated in mean square convergence. Second, the corresponding unknown inputs are inhibited by using output errors of the estimator. In addition, the corresponding event-driven transmission strategy is designed by using a quadratic performance index, which guarantees a good balance between the estimation error and the data transmission rate as well as prolonged service life of the sensor battery. Finally, numerical simulation tests verify that the designed event-driven state estimator can estimate the system's state effectively and extend the sensor's battery life by approximately 48%. The proposed algorithm also leads to reduced utilization of network resources to some degree.
在本研究中，为包含未知输入和过程以及相关测量噪声的随机系统设计了一种事件驱动状态估计器。首先，利用随机稳定性理论和李雅普诺夫函数推导事件触发状态估计器的增益。然后，基于这些结果，计算均方收敛下的相应状态估计误差。其次，利用估计器的输出误差抑制相应的未知输入。此外，通过使用二次性能指标设计相应的事件驱动传输策略，这保证了估计误差与数据传输速率之间的良好平衡以及传感器电池的延长使用寿命。最后，数值模拟测试验证了所设计的事件驱动状态估计器能够有效地估计系统状态，并将传感器的电池寿命延长约48%。所提出的算法在一定程度上还降低了网络资源的利用率。
