Odometry based on auto-calibrating inertial measurement unit attached to the feet

D. Atchuthan, A. Santamaria-Navarro, N. Mansard, O. Stasse, J. Solà,
European Conference on Control, 2018, HAL Bib

Abstract:

Location of pedestrian in indoor environmentremains an open problem. A cheap and reliable sensor in this context is the inertial measurement units (IMU), carried by thepedestrian while he/she is walking. However, due to the bias ofboth the accelerometer and the gyroscope, integrating directly the inertial measurements leads to tremendous drift, as the stateof the system (position, orientation, velocity, bias) is not fully observable. In this paper, we consider the specific case wherean IMU is attached to one of the pedestrian’s feet. We exploit specific prior knowledges (i.e. the fact that the foot lands at zerovelocity on a horizontal plane) in order to make the full state ofthe IMU observable. The inertial measurements and these prio rknowledges are gathered in a graphical model (a factor graph),and are exploited to build a maximum-likelihood estimator. The technical difficulty is to handle the size of the graph such thatit is tractable in a limited time window, that we do by relying on the pre-integration technique. In that existing framework,our contributions are to reformulate the pre-integration methodusing quaternions while giving a simpler algebraic formulation, and to apply this method for estimating the human foot-poseduring walking. We validate these concepts on several long-range trajectories capture with human subject and comparethe results with ground-truth measurements (coming from a motion capture system) and previous results of the state of theart.