Walking speed and slope estimation using shank-mounted inertial measurement units

This paper studied the feasibility of walking speed and slope estimation using shank-mounted inertial measurement units, and proposed a real-time stride-by-stride walking speed and slope estimation method. This method processed acceleration and angular velocity measured by a shank-mounted inertia measurement unit and output stride-by-stride walking speed and slope information in real time. Instead of estimating gait parameters indirectly, this method integrated the transformed accelerations to calculate stride length, walking speed and slope in each gait cycle. The main challenge is to determine the integration initial values at the beginning of each gait cycle. By taking the advantage of inverted pendulum model of walking, we identified a appropriate way to segment a walking sequence into gait cycles such that all of the required initial values were available. Experimental results demonstrated that the proposed method could accurately estimate the walking speeds under different walking conditions. However, the method estimated the slope with a lower accuracy.