osition Control Using Acceleration- ased Identification and eedback With Unknown easurement Bias

position-command-following problem for asymptotically stable inear systems is considered. To account for modeling limitations, e assume that a model is not available. Instead, acceleration ata are used to construct a compliance (position-output) model, hich is subsequently used to design a position servo loop. Furhermore, we assume that the acceleration measurements obtained rom inertial sensors are biased. A subspace identification algoithm is used to identify the inertance (acceleration-output) odel, and the biased acceleration measurements are used by the osition-command-following controller, which is constructed usng linear quadratic Gaussian (LQG) techniques. DOI: 10.1115/1.2807180