Step length adaptation for walking assistance

Lower exoskeleton has gained considerable interests in rehabilitation and health care applications, since it has the potential to help the paraplegia patients to walk again. On the control of these lower exoskeletons, the predefined gait control methods are commonly used, which requires the pilot to adjust his/her movements to follow the predefined motion trajectories. In the meanwhile, the desired trajectories are predefined from healthy persons or extrapolated from clinical gait analysis (CGA) datasets. However, in individual walking assistance situations, the exoskeleton should have the ability to adapt with variant motions of the pilot, since the pilot will change his motion in different walking situations. This paper presents a novel step length adaptation method to adapt the pilot's motion for walking assistance exoskeletons. In this paper, the exoskeleton robot is model as a special Hybrid Human-Exoskeleton Agent (HHEA), which considers both the exoskeleton and the pilot. The Dynamic Movement Primitives (DMP) is utilized to model the exoskeleton gait trajectories dynamically, in which the relationship between the gait length and Center of Mass (CoM) of HHEA is considered. In the training process, a Reinforcement Learning (RL) method is employed to update the parameters of dynamic gait model online. We demonstrate the efficiency of the proposed step length adaptation method in simulation environment as well as a lower limb exoskeleton system named as AssItive DEvice for paRaplegics (AIDER). Experimental results shows that the proposed step length adaptation method is able to adapt variant motions of the pilot during walking.