On uas sensor flight pattern determination for studies in atmospheric science

Mobile sensing of field quantities is a question of interest for underwater and atmospheric fields. Consequently a number of approaches have been recently developed. In this work, a covariance-based trajectory optimization approach was adopted to develop a method for quantified comparison of candidate flight trajectories for measurements of an atmospheric field. With this method, comparisons are also possible among logistics scenarios involving different numbers and types of UAS platforms. In this approach a posteriori error variance based on spatial and temporal decorrelation scales leads to the optimization integral that is numerically evaluated for each proposed trajectory. A CM1 convective boundary layer simulation dataset was used for the initial evaluation of the method and to determine different spatial decorrelation scales in different directions. In addition, a dataset from a fixed weather tower was used to understand the impact of the temporal decorrelation scale. Ultimately, the uncertainty of field quantity estimates resulting from trajectory families were compared to determine which candidate trajectories are more suitable or are comparable for sampling the field variables of interest.