The nonlinearity and universal approximation property of fuzzy scheme have made it ideal tool to handle these situations. However, the knowledge about the system’s property plays a very important role in fuzzy scheme and in many cases this knowledge is so poor or even doesn’t exist. Combining fuzzy control with other effective theories like robust and adaptive schemes could make it independent of the control designer’s knowledge.In this dissertation, two novel schemes are developed by combination of adaptive method (section4) and robust adaptive method (section5) with fuzzy algorithm for tracking problems to guarantee uniform stability of the whole system. Adaptive fuzzy control (AFC) approach is developed in the presence of parametric uncertainties and unknown external disturbances. In robust adaptive fuzzy (RAFC) approach the effect of unmodelled dynamics also added. It is assumed that every external disturbance is the direct effect of environmental properties. In fact, it is resulted by the reaction between environment and manipulator motion. These disturbances can be modeled as nonlinear functions of robot links positions and velocities. The fuzzy part with a set of tunable parameters is used to approximate and reject the effects of external disturbances. Two adaption laws are introduced to estimate unknown parameters and determine an upper bound of unmodelled dynamics . A Lyapunov-based proof is presented to guarantee uniform stability of the proposed RAFC system. The simulation results show the excellent control performance of the proposed approach to tracking the desired trajectory and its capability to reject the effect of unknown force/torque disturbances and unmodelled dynamics. The robustness of the system has been ensured by removing the effect of unmodelled dynamics and estimation error of the ideal fuzzy parameters with finding an upper bound of them. The proposed AFC and RAFC controls have used fuzzy scheme, based on nonlinearity properties to model the external disturbances. The dynamic fuzzy parameters have made fuzzy estimator powerful to reject the effect of unknown environmental forces or torques. To avoid chattering problems the control laws have been modified without losing any stability conditions. It has been also proved that the tracking errors are converge to zero as time goes to infinity, estimations are always staying bounded, and uniform stability is independence of initial values. One of the greet advantage of the proposed algorithms is that the control designer has no problem to construct the fuzzy rule base because the rules are self-constructive. Finally as shown in simulations, the novel AFC and RAFC are robust to parameter variations, unmodelled dynamics, and external disturbances. Keywords: Adaptive fuzzy control, Robust control, Trajectory tracking, Robot manipulator, disturbance rejection.