摘要: |
An ad hoc unmanned ground vehicle (UGV) network operates as an intermittently connected mobile delay tolerant network (DTN). The path planning strategy in a DTN requires mobility estimation of the spatial positions of the nodes as a function of time. The purpose of this thesis is to create a foundational mobility estimation algorithm that can be coupled with a cooperative communication routing algorithm to provide a basis for real time path planning in UGV-DTNs. In this thesis, we use a Gauss-Markov state space model for the node dynamics. The measurements are constant power received signal strength indicator (RSSI) signals transmitted from fixed position base stations. An extended Kalman filter (EKF) is derived for estimating of coordinates in a two-dimensional spatial grid environment. Simulation studies are conducted to test and validate the models and estimation algorithms. We simulate a single mobile node traveling along a trajectory that includes abrupt maneuvers. Estimation performance is measured using zero mean whiteness tests on the innovations sequences, root mean squared error (RMSE) of the state estimates, weighted sum squared residuals (WSSRs) on the innovations, and the posterior Cramer-Rao lower bound (PCRLB). Under these performance indices, we demonstrate that the mobility estimator performs effectively. |