基于衰减信道的几类随机非线性系统的模糊滤波方法研究

The filtering problem for nonlinear stochastic systems with randomly occurred incomplete information is still a very valuable research topic. Especially under the channel fading situation, the performance analysis and design of the filter based on the theory of the fuzzy system is more challenging, is a frontier research topic in the field of control. This project is devoted to investigating the multi-objective analysis and filter problems of stochastic fuzzy systems based on the fading channel of wireless networks. First of all, according to the channel fading phenomenon which commonly exists in the wireless communication networks, the causes and property of channel fading are comprehensively analyzed. Then, a new reasonable measurement system of nonlinear stochastic dynamic model and measurement model is established combining with various types of randomly occurring incomplete phenomena (such as missing measurements, distributed time-delays, gain variations and so on). The T-S fuzzy theory is used to analyze the system of the stability, the robustness and the non-fragility and other performance. Furthermore, the expected fuzzy filter structures and parameters are also proposed which are convenient to implement in the practical engineering, and the existence criterion and optimal fuzzy filtering algorithm of desired controller are derived simultaneously. Finally, the obtained results will be verified in the wireless mobile robot localization experiment. This project will greatly enrich the existing research on wireless communication network, will try to improve and complete the fuzzy filtering theoretical system from theoretical and technical aspects. Therefore, the research results have high theoretical value and broad application prospects.

带有随机发生信息不完整的随机非线性系统的滤波技术研究仍是一个热点问题。特别在信道衰减情形下，基于模糊系统理论的滤波器性能分析与设计更具挑战性，是控制领域的前沿研究课题。本项目将以无线通信网络中普遍存在的信道衰减现象为切入点，全面分析信道衰减的本质特征及成因，并综合考虑信息传输中的各种随机发生信息不完整现象(如信号丢失现象、分布时滞现象及传感器饱和现象等)，建立相应的测量系统的随机非线性动力学模型与测量模型。在此基础上，应用T-S模糊理论分析滤波系统的稳定性、鲁棒性以及非脆弱性等多种性能。进而提出易于工程实现的模糊滤波器结构与参数设计方案，获得满足给定性能指标的模糊滤波器存在性条件及优化算法。最后，将成果在无线网络下移动机器人定位实验中进行验证。本项目将大大地丰富、拓展现有随机非线性滤波的研究成果，力图从理论与技术层面上完善模糊滤波的理论体系，其研究成果具有较高的理论价值和广泛的应用前景。

带有随机发生信息不完整的随机非线性系统的滤波技术研究仍是一个热点问题。特别在信道衰减情形下，基于模糊系统理论的滤波器性能分析与设计更具挑战性，是控制领域的前沿研究课题。本项目将以无线通信网络中普遍存在的信道衰减现象为切入点，全面分析信道衰减的本质特征及成因，并综合考虑信息传输中的各种随机发生信息不完整现象(如信号丢失现象、分布时滞现象及传感器饱和现象等)，建立相应现实通信环境下的测量系统的随机非线性动力学模型与符合实际情况的测量模型。在此基础上，应用T-S模糊理论分析滤波系统的稳定性、鲁棒性、非脆弱性H2性能和H∞性能等多种性能。进而提出易于工程实现的模糊滤波器结构与参数设计方案，获得满足给定性能指标的模糊滤波器存在性条件及优化算法，完善和评价所得研究结果与算法。此外，应用人工神经网络方法，针对一种同时存在多重衰减和冗余通道的人工神经网络和忆阻神经网络，提出了H∞状态估计器的新设计问题。 从工程的角度来看，这些被采用的冗余通道可以有效地提高网络系统的可靠性。最后，将成果在无线网络下移动机器人定位实验中进行验证。本项目将大大地丰富、拓展现有随机非线性滤波的研究成果，力图从理论与技术层面上完善模糊滤波的理论体系，其研究成果具有较高的理论价值和广泛的应用前景。