This paper presents a data-driven control design framework to achieve robust tracking control without exploiting mathematical model of nonlinear underactuated mechanical systems (UMS). The method leverages the differential flatness property of linearized systems and online estimation and compensation of disturbances by active disturbance rejection control (ADRC). The differentially flat output is derived directly from measured data with unknown dynamics and parameters of UMS by the flat output identification (FOID) algorithm. A reduced nominal model of UMS is proposed to simplify the process of finding flat output and trajectory planning. Technique of sparse regression is applied to identify the relationships between flat output and system states, which reduces the order of the well-known extended state observer (ESO) and thereby make the ESO more effective for both trajectory planning and tracking in terms …