Task-dependent modulation of multi-dimensional human ankle stiffness

While recent research on controlling an ankle joint of the lower extremity robots that mimic intact human ankle stiffness has gained much attention, this approach is still limited by the fact that modulation of human ankle stiffness is not fully understood yet. This chapter presents a series of our recent efforts to quantify two-dimensional ankle stiffness during various motor tasks including seated, standing, and walking tasks. Robotic devices capable of actuating 2 DOF (degrees-of-freedom) of the ankle, in particular, dorsiflexion-plantarflexion in the sagittal plane and inversion-eversion in the frontal plane, in combination with system identification methods, allowed us to reliably quantify two-dimensional ankle stiffness during static and dynamic tasks that involve a wide range of muscle activation patterns. Our studies have demonstrated that human ankle stiffness modulates in a task-dependent manner, but it is consistently greater in the sagittal plane than in the frontal plane. The studies have also confirmed that quantification results in one specific task cannot be directly translated to explain ankle stiffness modulation in other tasks, emphasizing the importance of quantification over various motor tasks. Ultimately, integration of findings observed from a wide range of tasks will allow us to construct a universal ankle stiffness model that better explains task-dependent modulation of multi-dimensional ankle stiffness and provides more reliable information to robotic controllers.