In this paper a feedback linearization (FL) and extended state observer (ESO) based control law is proposed for the trajectory tracking control of a flexible joint robotic system. The FL based controller requires the transformed full state vector for its implementation. The design also requires exact knowledge of the system model making the controller performance sensitive to uncertainties. To address these issues, an ESO is designed which estimates the state vector as well as the uncertainties in an integrated manner. The FL controller uses the states estimated by ESO and thus the issue of availability of the states is addressed. Further the effect of uncertainties is compensated by augmenting the FL controller with the ESO estimated uncertainties. The effectiveness of the ESO in estimation of the uncertainties and states and tracking performance of the augmented controller are demonstrated through simulation.