Cooperative target localization by multiple unmanned aircraft systems using sensor fusion quality

Over the past several years, researchers at the U.S. Air Force Academy developed cooperative, distributed aerial sensor networks (Pack et al. in IEEE Trans Syst Man Cybern Part B Cybern 39(4):959–970, 2009) using multiple small unmanned aerial vehicles (SUAVs) to search, detect, and locate ground targets. The use of distributed SUAVs, however, introduced a set of problems, including difficulties in reliable air‐to‐air communication and clock synchronization among onboard systems of multiple SUAVs. The communication problems further aggravate the synchronization problem contributing to a large target localization error. Conventional methods use multiple sensor outputs of the same target seen from different perspectives to increase the target localization accuracy. These methods are effective only when the pose errors of sensor platforms based on GPS data are modeled accurately, which is not a reasonable assumption for SUAVs, especially when SUAVs operate in an environment with wind gusts. In this paper, we propose a robust, novel technique that analyzes what we call ‘sensor fusion quality’ to assign an appropriate sensor reliability value to each set of updated sensor data. In the proposed approach, we characterize the quality of a set of newly acquired sensor data, containing a target, by examining the joint target location probability density function. The validity of the proposed method is demonstrated using flight test data.