In the filling phase of injection molding process, melt-front rate plays a particularly important role in determining the part quality. Prior researches recommend a constant melt-front rate for this phase. A measurement device based on a capacitor principle has been developed. It has been shown that this sensor is effective to measure online and continuously melt-front position in mold cavity. Based on this sensor, this work develops and tests closed-loop control system of direct in-mold melt-front rate control. Several advanced control algorithms, Generalized Predictive Control (GPC), Iterative Learning Control (ILC), and Generalized 2D Model Predictive Iterative Learning Control (2D-GPILC), are designed, implemented, and tested with different mold inserts. The results show that 2D-GPILC, a combination scheme of feedback and iterative learning scheme, gives the best control results.