GES source localization and navigation based on discrete-time bearing measurements

Abstract

This paper addresses the problems of source localization and navigation based on discrete-time single direction measurements, in 3-D, in addition to relative velocity readings. For source localization, an agent aims at estimating the position of a source relative to itself, while for navigation the agents aims to estimate its own position assuming that it also has access to the position of the source in an inertial frame. Additionally, unknown constant drift velocities are considered and explicitly estimated. The design follows essentially by considering an augmented system, which is linear, and thoroughly address its observability and its relation with the original nonlinear system. The final estimation solution is a Kalman filter, with globally exponentially stable (GES) error dynamics. Simulation results are presented that illustrate the achievable performance with the proposed solution.