The ability to walk without the help of a caretaker enhances the quality of life for those who are bed-ridden or confined to a wheelchair. At present, most of the available gait rehabilitation robot systems have been designed to support the body weight externally. For gait training to be effective, a mobile body weight support (BWS) mechanism is needed. In mobile gait training robot systems, functions such as patient path following and constant BWS are important issues, particularly in dynamic environments. In the present study, two types of robotic systems were developed for gait rehabilitation. The first is known as the mobile manipulator type and the second the mobile vehicle type. The differences between the two systems in design and control are discussed. A control algorithm based on a neural network was used to compensate for dynamic interactions, unmodeled dynamics, and disturbances by the user on the system. Both electrical and pneumatic BWS mechanisms were built and compared. The proposed BWS systems were tested experimentally for their effectiveness in gait rehabilitation while maximizing the therapeutic outcome.