Fusion-based localization for a Heterogeneous camera network

Heterogeneous sensor networks (HSNs) are becoming more commonly used for purposes such as monitoring and surveillance, as they offer richer sources of data for situational awareness. An important aspect of HSNs is localization. In this paper, we describe a novel method for localizing a network of cameras equipped with wireless radios. Our method fuses both the image data and radio interferometry data in order to determine the position of the sensors and the orientation of each camerapsilas field of view. While existing methods that rely solely on image data alone are often limited in that they can only recover position up to scale factors, by fusing the image data and radio interferometry data, we are able to recover the position and orientation with no scale factor ambiguity. In contrast, localization of sensor nodes using radio alone only recovers the position of the sensors and often relies on computationally expensive methods. The method discussed in this paper exploits both the image and radio data for a more computationally efficient process of localization. We discuss both a linear and nonlinear approach to fusing the data which depend on different constraints on the network. We demonstrate our approach on a real network of camera and radio nodes.