This idea makes use of Adaptive Neural-Network Predictive Controllers (ANNPC) in conjunction with GPS signals to control the orbit and attitude of any type of Earth-orbiting spacecraft. The simulation models we have developed demonstrate that one can implement an orbital control system for spacecraft by combining ANNPC with input state vectors generated from GPS signals received on board. The key advantage of using ANNPC is that it does not require highly accurate and costly dynamic models for specific spacecraft to enable orbital and attitude prediction and control for every new spacecraft design. Instead, a generic ANNPC algorithm can be developed and trained to learn the orbital and AOCS dynamics of spacecraft during their preoperational and operational phases. The simulations have demonstrated that using such a system optimizes spacecraft thrust forces, thus reducing fuel consumption and prolonging missions by more than 30%. By using ANNPC-GPS, it is possible to reduce, or even eliminate, the reliance on ground control station (GCS) telemetry and ranging and tracking antenna (TTAC) systems (TTAC accounts for up to 50% of GCS costs).