In this paper, a novel robust adaptive control design for a general class of Single-Input Single-Output (SISO) nonlinear time-varying systems is delineated. To this end, for a ball handling mechanism, a systematic approach is presented to modify solutions of the arm target angle tracking problem in this robot. Initially, a virtual model of simplified ball handling mechanism is sketched in MSC-ADAMS. Afterwards, assuming there is no priori knowledge about the dynamics of the system except its order, a brand new suggested online linearization technique based on Recursive Least Squares (RLS) method is applied inorder to identify the plant and to derive an Adaptive Linear Regression (ALR) model for the system. Concentrating on this pseudo linear form, an extended sliding mode controller, established upon Lyapanuv theory, is designed and developed step by step. Multifarious simulations were performed and results clearly demonstrated the privilege and effectiveness of the proposed approaches.