This paper describes a control system which can provide the assistive torque for the lower limb exoskeleton. To provide the assistive torque for the lower limb exoskeleton need to be known the gait phases in one gait cycle walking as the reference signal. The gait cycle information already got from another work with the same prototype of the exoskeleton. The gait cycle signal has been generated from Inertial Moment Unit (IMU) sensor and built-in encoder on the Maxon motor. The decision tree method has been used as the proposed method in order to provide the suitable assistive torque. The real-time experimental result has been carried out to verify the performance of the proposed method. To examine the proposed method in a real-time application, some walking type has been done. The result shows that the proposed method can provide the assistive torque even in several different walking types.