A Class of Robust Non-linear Extended State Estimation with Colored Noise
A robust non-linear extended state estimation algorithm is designed for networked sys-tems containing coloured noise and limited communication bandwidth.The measurement information method of differences is used to reduce the influence of noise correlation on the system,and then the logarithmic quantizer is used to process the measurement information to overcome the communication capacity limitation,the uncertain non-linear item is innovatively regarded as a state of extended state to solve the non-linear problems.On the basis,a robust nonlinear extended state estimator is designed.Using Young's inequality and matrix analysis are applied to solve the Riccati equation,the upper bound of the error covariance of the estimator is obtained,and the real-time optimization of the gain is real-ized.The estimation algorithm is more convenient to deal with non-linear systems,and it can overcome the conservativenessin the treatment of non-linear functions,and has excellent estimation performance.
state estimationnon-linear time-varying systemextended statequantization processrobust control