The demand for walking training robots has increased owing to a serious shortage of rehabilitation physiotherapists. However, the intelligence level of existing rehabilitation training robots is low; these robots cannot realize a direct switch between passive and active training, which makes rehabilitees lack interest in training and deteriorates the effect of rehabilitation training. In this study, a rehabilitation gait training robot is developed in line with the characteristics of human omnidirectional walking. The proposed robot uses a suitable control algorithm to accurately follow the exercise programs prescribed by physical therapists and can realize fine practice results. The novelty of the robot is that passive and active training can be directly and gently switched during walking. The passive training stage has a velocity restriction safety function, whereas the active training stage has a velocity decision function. The purpose of this design is to avoid sudden changes in the robot velocity during the passive training stage and to guarantee the coordination of the human–robot velocity in the active training stage. Comparative simulation analyses and experimental results show that the proposed passive and active direct switching training improves the intelligence and security of rehabilitation robot.
