An Adaptive Shared Control of a Novel Robotic Walker for Gait Rehabilitation of Stroke Patients

An increasing number of stroke patients are suffering from motor disability. For enhancing their quality of daily life, recovery from disabled ambulation is in urgent need. Traditional robotic gait rehabilitation systems with treadmill assistance constrain stroke patients in sagittal plane and restricting pelvic horizontal motion. In order to solve these problems, A novel robotic walker was proposed for effective gait rehabilitation. It is composed of a body weight support (BWS) unit and an active omni-directional mobile platform to assist gait training, including anterior-posterior, medio-lateral, rotational pelvic movements and supports for balance and postural control. In order to improve self-motivated ambulatory ability, an adaptive shared control strategy is proposed, which automatically executes human-in-charge or robot-in-charge mode during gait training. 3 healthy young subjects were recruited in experiments to assess performance and effectiveness of this controller. Their kinematic changes and muscle activities in experiments showed beneficial influence of the robotic walker with the adaptive shared control for gait rehabilitation.