A Robust Cooperative Estimation Approach for Unknown MIMO Nonlinear Systems With Unknown Disturbances

This article proposes a robust cooperative estimation method for discrete-time multi-input multioutput (MIMO) nonaffine nonlinear systems with unknown disturbances. It develops a standardized discrete state-space representation based on diffeomorphic transformation and proposes a discrete-time estimation framework to ensure inherent compatibility with digital implementation. The method integrates adaptive unknown disturbance, parameter, and state estimators into a synergistic framework that enables mutual correction and iterative updating, effectively overcoming the error accumulation inherent in traditional cascaded estimation architectures. High-gain techniques are incorporated into the subsystems, and rigorous Lyapunov stability analysis guarantees superior robustness and convergence against highly nonlinear dynamics and unknown disturbances. Theoretical results are validated through numerical simulations and a practical application to quadrotor dynamic estimation. Comparative studies with existing nonlinear estimators demonstrate the significant advantages of the proposed algorithm in estimation accuracy and stability. Specifically, under extensive robust testing scenarios, including various payload configurations and sustained wind disturbances, the proposed method achieves comprehensive average reductions in the overall estimation mean absolute error of 71.3% and 31.0% compared to the standard nonlinear adaptive observer and the cascaded method, respectively, alongside corresponding root-mean-square error reductions of 48.5% and 13.1%.