English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We address the problem of robot localization using ground penetrating radar(GPR) sensors. Current approaches for localization with GPR sensors require apriori maps of the system's environment as well as access to approximate globalpositioning (GPS) during operation. In this paper, we propose a novel,real-time GPR-based localization system for unknown and GPS-deniedenvironments. We model the localization problem as an inference over a factorgraph. Our approach combines 1D single-channel GPR measurements to form 2Dimage submaps. To use these GPR images in the graph, we need sensor models thatcan map noisy, high-dimensional image measurements into the state space. Theseare challenging to obtain a priori since image generation has a complexdependency on subsurface composition and radar physics, which itself varieswith sensors and variations in subsurface electromagnetic properties. Our keyidea is to instead learn relative sensor models directly from GPR data that mapnon-sequential GPR image pairs to relative robot motion. These models areincorporated as factors within the factor graph with relative motionpredictions correcting for accumulated drift in the position estimates. Wedemonstrate our approach over datasets collected across multiple locationsusing a custom designed experimental rig. We show reliable, real-timelocalization using only GPR and odometry measurements for varying trajectoriesin three distinct GPS-denied environments. For our supplementary video, seehttps://youtu.be/HXXgdTJzqyw.

Ground Encoding: Learned Factor Graph-based Models for Localizing Ground Penetrating Radar