Many research groups are developing robotic exoskeletons which aim to assist motor rehabilitation for individuals after neurological injuries. As a rehabilitation tool, the primary goals of robotic exoskeletons are to reduce manual assistance from therapists and to achieve optimal training outcomes. Although the first objective has been realized by many current robotic devices, producing better rehabilitation outcomes with robotic devices is still a developing area of research. To design better robotic devices, it is important to understand the principles governing how humans learn to interact with the robotic assistance and to identify the gait parameters humans prioritize as objectives for their gait pattern. I used two types of robotic exoskeletons to examine rapid locomotor adaptation to mechanical assistance. The specific questions I addressed were: (1) how do neurologically intact subjects adapt their walking patterns to a powered orthosis that provides dorsiflexion assistance? (2) Do subjects walk with joint kinetic patterns during robotic-assisted walking similar to joint kinetic patterns during unassisted walking? (3) Is locomotor adaptation rate dependent on exoskeleton strength? (4) Is an increased stretch reflex inhibition a potential mechanism for reducing soleus recruitment when walking with the exoskeleton? (5) Do stretch reflex responses during unexpected gait perturbations appropriately meet the mechanical requirements of gait? In the first experiment, the exoskeleton was configured to provide dorsiflexion assistance proportional to the voluntary activation of the tibialis anterior muscle (i.e., ankle dorsiflexor). For the rest of experiments, the exoskeleton was configured with two artificial pneumatic muscles providing plantar flexion assistance proportional to the soleus muscle activation (i.e., ankle plantar flexor). The double muscle exoskeleton provided twice the mechanical capability of the previous single-muscle exoskeleton design (Cain et al. 2007; Gordon and Ferris 2007; Kinnaird and Ferris 2009). The results provide important insights into the principles governing human locomotor adaptation. The information can be used to aid in designing powered devices and may be helpful in optimizing gait rehabilitation therapies. In addition, the findings should be helpful in the development of biologically realistic neuromusculoskeletal computer simulations of human walking.

Категория: Научные статьи | Нет комментариев »

Комментарии