Adaptive observer design based on an algebraic modelling approach

This paper studies the problem of designing adaptive observers for linear time-invariant systems with unknown inputs when the observer matching condition is not satisfied. The unknown inputs are divided into two parts: one is matched and the other is mismatched. Under the assumption that the mismatched unknown inputs are deterministic and can be expressed as the response of a fictitious autonomous system, an augmented system consisting of the original system and the mismatched unknown inputs dynamical system is constructed so that the observer matching condition can be satisfied. For the augmented system, the procedure to design this adaptive observer is given, and it is able to not only observe the state of original system but also reconstruct the matched unknown inputs. Such adaptive observer is applied to a kind of robot system and the simulation results demonstrate the effectiveness of the proposed approach. ©, 5015, Xi'an Jiaotong University. All right reserved.