The objective of this study is to design an optimal model predictive control (MPC) strategy using manipulated variables to control the production of a sufficient amount of hydrogen through low temperature autothermal reforming of dimethyl ether (DME), a reforming reaction performed using PdO/ZnO/γ-Al2O3 catalysts coated on honeycomb cordierite ceramics. Experiments and simulation have verified that the optimal activity temperature of the catalyst is 400 °C, and the hydrogen volume fraction in syngas is over 43%. In the implementation of the hydrogen production system, the MPC controller can precisely determine the feed rates of DME, high-purity air, and water based on the space state equation of the reformer, to achieve the anti-disturbance of the reformer temperature. Thus, the reduction of hydrogen yield and sintering of the catalyst as a result of overheating are prevented. As the static and dynamic performance of hydrogen production exhibits excellent tracking of the setpoints, an autonomous, automated, and reliable continuous system was designed to meet the desired hydrogen demand situation. This study shows that an autonomous, automated, and reliable continuous hydrogen production reforming system can be designed to actively respond to the on-board hydrogen usage situation.