Distributed extended high-gain observers for the generalized strict-feedback system

Two distributed extended high-gain observers are studied for the generalized strict-feedback system in presence of external disturbance. The system state dynamics are reconstructed admissibly based on the limited measured output. First, a distributed high-gain observer is constructed to ensure the semiglobal ultimate uniform convergence of the state estimation error between the estimated state in each local observer and system state, for any prescribed initial system values. However, the dimension of the proposed distributed observer is n + 1 and the high-gain parameter is powered up to n + 1. Therefore, a 2(n + 1)-dimension distributed high-gain low-power observer is delicately designed, with the high-gain parameter powered fixed to 2. The formulated distributed observer is easy-implementation and the estimation error can reach fast convergence. Finally, a chaotic oscillator and a pendulum-cart system are introduced to verify the effectiveness of the theoretical results.

Automated summary

Two distributed extended high-gain observers are proposed for the generalized strict-feedback system with external disturbance. The state dynamics are reconstructed based on limited measured output. The first observer ensures the semiglobal ultimate uniform convergence of the state estimation error, while the second observer is designed to be low-power yet effective. The effectiveness of the proposed observers is verified through experiments on a chaotic oscillator and a pendulum-cart system.