Development aspects of a robotised gait trainer for neurological rehabilitation

The restoration of gait is a key goal after stroke, traumatic brain injury and spinal cord injury. Conventional training methods, e.g. treadmill training, require great physical effort from the therapists to assist the patient. After the successful development and application of a mechanised gait trainer, a new research project of constructing a sensorised robot gait trainer is under way. The aim of this project is to build a robotic device which enables the therapist to let the machine move the patients feet, fixed on two footplates, on programmable foot trajectories (e.g. walking on the ground, stepping stairs up and down, disturbances during walking). Furthermore impedance control algorithms will be incorporated for online adaptation of the foot trajectories to the patients walking capabilities. Another important feature is the compliance control to simulate virtual ground conditions, i.e. the machine acts as a haptic foot device. Due to the partially high dynamic foot movements during normal walking, conventional industrial robots are not suitable for this task. This paper describes development aspects and problems that have to be dealt with during the design process of the robotised gait training machine.