论文部分内容阅读
对于带不确定模型参数和噪声方差的线性离散时不变多传感器系统,用虚拟噪声补偿不确定参数,系统转化为仅带噪声方差不确定性的多传感器系统.用加权最小二乘法和极大极小鲁棒估计准则,基于带噪声方差保守上界的最坏情形保守系统,提出一种鲁棒加权观测融合稳态Kalman预报器,并应用Lyapunov方程方法证明了它的鲁棒性,同时给出了与鲁棒局部和集中式融合Kalman预报器的精度比较.最后通过一个仿真例子说明了如何搜索参数扰动的鲁棒域,并验证了所提出的理论结果的正确性和有效性.
For uncertain discrete-time invariant multisensor systems with uncertain model parameters and noise variance, the system is transformed into a multisensor system with only noise variance uncertainty by using the uncertain parameters of virtual noise compensation.Using the weighted least square method and the maximum Based on the worst conservative system with conservative upper bound of noise variance, this paper proposes a robust weighted observation fusion steady-state Kalman predictor, and proves its robustness by using Lyapunov equation Finally, a simulation example is given to illustrate how to search the robust domains perturbed by parameters and verify the correctness and validity of the proposed theoretical results.