Robot modeling is essential before using control systems to verify that the desired task is completed with the lowest feasible error rate according to the inputs. When modeling a robot, the derivation of the forward movement of the robot axes is a fundamental step based on the method of Denavit–Hartenberg. This research aims to control the robot arm motion using two controllers. PID controller was used as a reference to compare the results with those of an FLC. First, an FLC was used to enhance the nonlinearity of the robot arm.FLC was designed based on Mamadani pro-Max inference. four different defuzzification methods were used and compared to obtain the control signal. These are BOA, MOM, SOM, and COG.The results of this controller were then compared to the PID results on a transient response scale. Finally, simulation was done using MATLAB Simulink software.Based on the simulation results, BOA, MOM, and SOM techniques yield almost identical results; however, the COG strategy yields a broad range of outcomes. Implementing a simple defuzzification approach resulted in system optimization due to the complexity of processes like fuzzification and defuzzification.Based on the simulation results, the FLC system yields better results than those obtained with a PID. The FLC has lower rise time, settling time, steady-state error, and less overshoot than the PID controller.