Underground Mapping and Localization Based on Ground-Penetrating Radar

Abstract

3D object reconstruction based on deep neural networks has gained increasing attention in recent years. However, 3D reconstruction of underground objects to generate point cloud maps remains a challenge. Ground Penetrating Radar (GPR) is one of the most powerful and extensively used tools for detecting and locating underground objects such as plant root systems and pipelines, with its cost-effectiveness and continuously evolving technology. This paper introduces a parabolic signal detection network based on deep convolutional neural networks, utilizing B-scan images from GPR sensors. The detected keypoints can aid in accurately fitting parabolic curves used to interpret the original GPR B-scan images as cross-sections of the object model. Additionally, a multi-task point cloud network was designed to perform both point cloud segmentation and completion simultaneously, filling in sparse point cloud maps. For unknown locations, GPR A-scan data can be used to match corresponding A-scan data in the constructed map, pinpointing the position to verify the accuracy of the map construction by the model. Experimental results demonstrate the effectiveness of our method.

Figures (8)

• Figure 1: GPR sensing principles. Left: GPR emits EM radio wave pulses into the ground via an antenna. Pulses reflect back when hitting objects with different electromagnetic properties. Right: Multiple B-scans along the moving direction can represent the sparse 3D underground object shape.
• Figure 2: Omnidirectional robot for GPR data collection, where the robot (HK1500) tows a GPR cart (Leica DS2000).
• Figure 3: Parnet uses YOLOv8 architecture trained using the multi-task loss to map an GPR Bscan image to a set of output grids $\hat{\mathbf{G}}$ containing the predicted keypoints and bounding boxes. The equation of a hyperbola can be derived from its keypoints.
• Figure 4: The GPRNet multi-task framework takes sparse point cloud data as input and simultaneously outputs segmentation results and the completed point cloud.
• Figure 5: The GPRNet multi-task framework takes sparse point cloud data as input and simultaneously outputs segmentation results and the completed point clouds.