A new model reference adaptive control with the disturbance observer-based adaptation law for the Nonlinear Servomechanisms: SISO and MIMO Systems

A disturbance observer (DOB) is proposed to replace the external disturbance adaptation law in model reference adaptive control (MRAC) to perform robust tracking control performance for nonlinear servomechanisms. Lyapunov stable theory is implemented to combine the DOB with MRAC to simultaneously converge the estimated states and tracking errors toward null. Then, the disturbance observer-based MRAC (DOBMRAC) is proposed and realized. The single-input single-output (SISO) mass-spring-damper system and multi-input multi-output (MIMO) two-link manipulator robot system are both demonstrated as the control examples. However, in comparisons of tracking control performance between MRAC and DOBMRAC, it is found that the proposed controller reveals better robust tracking control performance for SISO and MIMO servomechanisms with unknown time-varying disturbances. The contributions of this paper are: i) the proposed DOBMRAC is successfully implemented in both the SISO and MIMO systems; ii) the proposed controller performs better tracking control performance than MRAC for unknown time-varying disturbances; iii) the DOBMRAC cannot only perform robust tracking control performance, but also shows disturbance estimation ability well; and iv) the proposed control methodology can also be widely applied to other MRAC systems with unknown time-varying disturbance.

Automated summary

This study proposes a disturbance observer (DOB) to replace the external disturbance adaptation law in model reference adaptive control (MRAC) for robust tracking control performance in nonlinear servomechanisms. The proposed DOBMRAC demonstrates better tracking control performance than MRAC for SISO and MIMO systems, while also showing good disturbance estimation ability. The methodology can be widely applied to other MRAC systems with unknown time-varying disturbance.