Assessment of Foot Kinematics During Steady State Running Using a Foot-mounted IMU

Abstract

The continuous sensing of kinematics provides an opportunity to monitor changes in sporting technique or to aid in injury rehabilitation. Inertial sensors are now small enough to integrate into footwear providing a potential platform for continuous monitoring that does not require additional components to be worn by the athlete. We investigate the suitability of using foot mounted inertial sensors to assess foot kinematics in steady state running by employing inertial navigation and pedestrian dead reckoning techniques. The results are evaluated by comparison with an optical motion capture system. Two techniques are assessed, the use of an Extended Kalman Filter with zero velocity updates and a linear de-drifting technique. Assessment of two example metrics, foot clearance (FC) and mean step velocity (SV), produced a bias and standard deviation of 0.000m±0.008 (FC) and 0.04m/s±0.03 (SV) using the linear de-drifting technique. Similar results were obtained using an Extended Kalman Filter approach calculating FC with a bias of 0.002m±0.029 and SV with a bias of 0.03m/s±0.02.