Loosely-Coupled GNSS/INS Integration for Foot Trajectory Reconstruction in Outdoor Environments

Many application scenarios, such as sports, well-being, and rehabilitation, would benefit from continuously recording human movements in outdoor, unstructured environments. A novel wearable device, which integrates an Inertial Measurement Unit (IMU) and a Global Navigation Satellite System (GNSS) receiver, can reconstruct foot trajectories using a loosely-coupled (LC) integration approach. In this approach, the IMU provides a raw position estimate through an inertial navigation system (INS) method, with zero-velocity updates when a stance phase is detected. This raw estimate is then corrected by GNSS-based position and velocity recordings. We tested the device in a field experiment where subjects walked at various speeds (slow, comfortable, fast) along a fixed path. For slow and comfortable walking speeds, the horizontal positioning error remained within 1 to 10 centimeters. At faster walking speeds, the error increased. We found that this increased error was primarily due to a reduction in the number of valid visible satellites observed during phases of the movement when the foot experienced larger accelerations, which led to lower precision in the GNSS position estimates. The impact of this effect is partially mitigated by rejecting low-quality GNSS corrections that correspond to a lower number of valid satellites.