A recursive least square approach to a disturbance observer design for balancing control of a single-wheel robot system

Disturbance observer(DOB) has been known as a simple robust control method to reject disturbances. Design of Q filters is considered as the most important design technique for the DOB design to compensate for the possible improperness of the inverse model of the plant as well as a non minimum phase. Many efforts on Q-filter design have been accomplished to realize a stable filter for the inverse of the nominal model in the DOB. In this paper, a recursive least square approach to a disturbance observer (RbDOB) for modeling the plant is proposed. A recursive least square method is used to estimate model parameters of the 2
^nd
order model through an input-output FIR filtering. Experimental studies of balancing control performances of a single-wheel robot by DOB and RbDOB are compared.