| Location information is of utmost importance for research and applications in both wireless sensor networks (WSNs) and mobile robotics. This work demonstrates the feasibility of combining two different location estimation techniques to improve the overall location and tracking capabilities of low complexity devices such as mobile sensor nodes in a WSN.;A Gaussian DR positioning error model is developed, and verified based on empirical measurements collected with a robot chassis. A Kalman filter, utilizing the Gaussian model is implemented to track the robot. The accuracy of the filter is compared with the Cramer-Rao lower bound (CRLB).;The system model is made up of stationary wireless devices known as beacons whose locations are known, and the low complexity device trying to know its position. The low complexity device localizes using the robot's dead reckoning (DR) position estimate and the received signal strength (RSS) range estimates made between itself and the beacons. |
