Localization and Mapping of Sparse Geologic Features with Unpiloted Aircraft Systems

Robotic mapping is attractive in many science applications that involve environmental surveys. This paper presents a system for localization and mapping of sparsely distributed surface features such as precariously balanced rocks (PBRs), whose geometric fragility (stability) parameters provide valuable information on earthquake processes. With geomorphology as the test domain, we carry out a lawnmower search pattern using an Unpiloted Aerial Vehicle (UAV) equipped with a GPS module, stereo camera, and onboard computers. Once a target is detected by a deep neural network, we track its bounding box in the image coordinates by applying a Kalman filter that fuses the deep learning detection with KLT tracking. The target is localized in world coordinates using depth filtering where a set of 3D points are filtered by object bounding boxes from different camera perspectives. The 3D points also provide a strong prior on target shape, which is used for UAV path planning to accurately map the target using RGBD SLAM. After target mapping, the UAS resumes the lawnmower search pattern to locate the next target. Our end goal is a real-time mapping methodology for sparsely distributed surface features on earth or on extraterrestrial surfaces.

Geomorphological Analysis Using Unpiloted Aircraft Systems, Structure from Motion, and Deep Learning

We present a pipeline for geomorphological analysis that uses structure from motion (SfM) and deep learning on close-range aerial imagery to estimate spatial distributions of rock traits (diameter and orientation), along a tectonic fault scarp. Unpiloted aircraft systems (UAS) have enabled acquisition of high-resolution imagery at close range, revolutionizing domains such as infrastructure inspection, precision agriculture, and disaster response. Our pipeline leverages UAS-based imagery to help scientists gain a better understanding of surface processes. Our pipeline uses SfM on aerial imagery to produce a georeferenced orthomosaic with 2 cm/pixel resolution. A human expert annotates rocks on a set of image tiles sampled from the orthomosaic, and these annotations are used to train a deep neural network to detect and segment individual rocks in the whole site. Our pipeline, in effect, automatically extracts semantic information (rock boundaries) on large volumes of unlabeled high-resolution aerial imagery, and subsequent structural analysis and shape descriptors result in estimates of rock diameter and orientation. We present results of our analysis on imagery collected along a fault scarp in the Volcanic Tablelands in eastern California. Although presented in the context of geology, our pipeline can be extended to a variety of morphological analysis tasks in other domains.