Dynamic modeling of human gait and motor adaptation: Towards the simulation of ankle impairments

Chad Everett Joshua MacDonald, University of Texas at El Paso


The simulation of gait, impairments and the outcomes of rehabilitation strategies is an important area of research. The ability to study the efficacy of treatment options before applying them to patients has the potential to improve the efficiency of the rehabilitation process. Any such simulations must incorporate some model of the motor learning process to properly anticipate the responses of patients. This dissertation presents initial work on the modeling of gait and the study of motor adaptation with a view towards simulating the motor adaptation process during gait. Specifically, an inverse dynamic model of gait incorporating sagittal plane limb and torso dynamics and three degree of freedom pelvis rotation is developed and applied to determine joint moments for motor adaptation experiments. The specific experimental conditions involve studying gait changes resulting from changes in treadmill walking speed, from walking on two treads moving at different speeds and various forms of ankle contractures simulated using an ankle-foot-orthosis. The experimental results show after-effects that suggest the formation of an internal model of the altered gait dynamics to compensate for gait errors. Kinematic performance metrics show the expected adaptations to the simulated gait impairments. In particular evidence is found to demonstrate vaulting and steppage gait adaptations in response to ankle impairments. The inverse dynamics analysis shows clear stance phase adaptations to ankle contractures while showing a limited response during swing phase. The primary adaptations evidenced by the inverse dynamics analysis are changes in pelvis joint moments leading to vaulting gait compensation when an ankle locked in an a plantarflexed position. The results of the adaptation experiments are compared to an inverse dynamics analysis of a patient exhibiting drop foot resulting from a mid-brain stroke. The patient also exhibits pelvis moments resulting a vaulting gait pattern as was seen in the plantarflexion experiment.

Subject Area

Biomedical engineering|Electrical engineering

Recommended Citation

MacDonald, Chad Everett Joshua, "Dynamic modeling of human gait and motor adaptation: Towards the simulation of ankle impairments" (2009). ETD Collection for University of Texas, El Paso. AAI3426849.