Adaptive-Resolution Field Mapping Using Gaussian Process Fusion With Integral Kernels

Unmanned aerial vehicles are rapidly gaining popularity in many environmental monitoring tasks. A prerequisite for their autonomous operation is the ability to perform efficient and accurate mapping online, given limited on-board resources constraining time computational capacity. To address this, we present an online adaptive-resolution approach field based on Gaussian Process fusion, a strategy which Bayesian fusion applied update prior map. key aspect of our integral kernel encoding spatial correlation over areas grid cells. This enables information compression uninteresting achieve compact map representation while maintaining correlations theoretically sound fashion. We evaluate performance both synthetic real-world data. Results show that method more terms memory consumption without compromising quality. Further, integrate into adaptive path planning framework it facilitates gathering efficiency settings.