Publication

Concurrent adjustment of active and passive optical sensors with GNSS and raw inertial data

2025

Abstract

This research focuses on augmenting the quality of digital mapping, notably from airborne sensors: camera and laser scanners. Progress is achieved by a joint combination of all critical observations in a common estimation scheme, called Dynamic Network (DN) or factor-graph. While earlier DN implementations treated imagery and lidar point-clouds independently, this thesis advances the methodology by integrating both datasets concurrently within a unified optimization framework. This research first investigates single-domain constraints, i.e. pixel-to-pixel (2D-2D) and point-to-point (3D-3D), in combination with satellite positioning (GNSS) and raw inertial observations. The methodology is then extended to cross-domain modeling by introducing a pixel-to-point (2D-3D) observation model, that enables direct coupling between images and point-clouds, starting with emulated data to establish theoretical observability. The extraction of 2Dâ 3D correspondences is realized first via domain-conversion techniques, i.e., 3D-3D links between dense image matching and lidar point-clouds or 2D-2D links between images and "rasterized" lidar clouds, and then via direct "learned methods". Results of single-domain constraints show considerable improvements in geo-referencing accuracy for inertial sensors of lower quality, as those used on smaller drones. The combined use of both 2D-2D and 3D-3D constraints improves the in-flight estimation of auxiliary parameters (as those related to system mounting - boresight), formally requiring specialized calibrations. Cross-domain conditioning shows great potential for further significant improvement of the geo-referencing accuracy when integrated with low-quality inertial sensors. Among the 2D-3D correspondence extraction techniques, those based on image and lidar clouds appear to be the most promising. The findings are supported by validations on data with modern sensors operated on a helicopter and an aircraft.

Official source

https://infoscience.epfl.ch/entities/publication/190609df-3b06-4a7b-bbdf-b6f414f47a63

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Graph Chat

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

2025

Abstract

Official source

https://infoscience.epfl.ch/entities/publication/190609df-3b06-4a7b-bbdf-b6f414f47a63

About this result

Related concepts (32)

Lidar

Lidar (ˈlaɪdɑːr, also LIDAR, LiDAR or LADAR, an acronym of "light detection and ranging" or "laser imaging, detection, and ranging") is a method for determining ranges by targeting an object or a surface with a laser and measuring the time for the reflected light to return to the receiver. LIDAR may operate in a fixed direction (e.g., vertical) or it may scan multiple directions, in which case it is known as LIDAR scanning or 3D laser scanning, a special combination of 3-D scanning and laser scanning.

Point cloud

A point cloud is a discrete set of data points in space. The points may represent a 3D shape or object. Each point position has its set of Cartesian coordinates (X, Y, Z). Point clouds are generally produced by 3D scanners or by photogrammetry software, which measure many points on the external surfaces of objects around them. As the output of 3D scanning processes, point clouds are used for many purposes, including to create 3D computer-aided design (CAD) models for manufactured parts, for metrology and quality inspection, and for a multitude of visualizing, animating, rendering, and mass customization applications.

3D scanning

3D scanner is the process of analyzing a real-world object or environment to collect three dimensional data of its shape and possibly its appearance (e.g. color). The collected data can then be used to construct digital 3D models. A 3D scanner can be based on many different technologies, each with its own limitations, advantages and costs. Many limitations in the kind of objects that can be digitised are still present. For example, optical technology may encounter many difficulties with dark, shiny, reflective or transparent objects.

Related publications (10)

Concurrent adjustment of active and passive optical sensors with GNSS and raw inertial data

Chat with Graph Search

FACE-ing the future of single-pixel complex-field microscopy beyond the visible spectrum

TopoTome: Topology-informed unsupervised segmentation and analysis of 3D images

Structure-aware Multi-view 3D Reconstruction of Dislocations in TEM with Message Passing Neural Networks