Nonlinear estimation of gait kinematics during functional electrical stimulation and orthosis-based walking

This paper presents a nonlinear estimation algorithm which utilizes a low-degree of freedom model of functional electrical stimulation (FES) and orthosis-based walking to estimate lower-limb angles. The estimated lower limb angles can be used to decide when the FES signal should be applied to the leg during the different phases of walking. To this end, we use measurements from inertial measurement units (IMUs) to estimate the lower limb segment angles. A state-dependent coefficient (SDC)-based nonlinear estimator is developed to estimate the lower limb angles. The nonlinear estimator is robust to uncertainties in the motion modeling and sensor noise/bias from the IMUs. A comparison with extended Kalman (EKF)-like filter shows improved performance of the estimator in simulation studies.