An Uncertainty-Driven and Observability-Based State Estimator for Nonholonomic Robots

The problem addressed in this article is the localization of a mobile robot using combination onboard sensors and ultrawideband (UWB) beacons. By discrete-time formulation system's kinematics, we identify geometric conditions that make system globally observable cast state estimation into framework least-squares optimization. observability filter thus obtained remarkably different from classic Bayesian filters, such as Kalman Filter, since it does not need priori stochastic models process measurement uncertainty contributions proves to have better performance than filters if are partly unknown or differ expected values. second important outcome work analytical study propagation. effectiveness designed filter, validity theoretical analysis uncertainties, comparisons with state-of-the-art extended (EKF) corroborated by extensive simulations validated experimentally.