Evaluating Wireless Network Effects for SLAM Robot Map Making

This work investigates how wireless network performance affects a rescue robot that is running a simultaneous localization and mapping (SLAM) algorithm. The SLAM algorithm is used to generate a map of an unknown disaster site. Odometry and laser range data collected by the robot is transferred via an ad-hoc wireless network to a central access point (AP) that is running the SLAM algorithm. Simulation is used to determine how losses, latency and the use of relays in the network affect the quality and completeness of the SLAM maps. This is illustrated using new metrics for SLAM map quality and completeness that are based on Hausdorff distance.