This paper introduces an online tuning method for a sliding mode fuzzy controller designed for robotic manipulators, aiming to enhance trajectory tracking while eliminating chattering without boundary layer saturation. The proposed controller uses a fuzzy inference system to adjust the sliding surface gain based on error inputs, ensuring stability and robustness in handling uncertainties. The effectiveness of the controller is validated through simulation results, demonstrating strong performance in terms of overshoot, rise time, and steady-state error.