Learning an Overlap-based Observation Model for 3D LiDAR Localization
This addresses localization for autonomous vehicles in urban settings, representing an incremental improvement over existing methods.
The paper tackles the problem of 3D LiDAR localization for mobile robots and autonomous cars by proposing a neural network-based observation model that predicts overlap and yaw angle offset between scans, and integrates it into a Monte-Carlo framework, achieving reliable localization in urban environments with superior global performance and fewer particles compared to baseline methods.
Localization is a crucial capability for mobile robots and autonomous cars. In this paper, we address learning an observation model for Monte-Carlo localization using 3D LiDAR data. We propose a novel, neural network-based observation model that computes the expected overlap of two 3D LiDAR scans. The model predicts the overlap and yaw angle offset between the current sensor reading and virtual frames generated from a pre-built map. We integrate this observation model into a Monte-Carlo localization framework and tested it on urban datasets collected with a car in different seasons. The experiments presented in this paper illustrate that our method can reliably localize a vehicle in typical urban environments. We furthermore provide comparisons to a beam-end point and a histogram-based method indicating a superior global localization performance of our method with fewer particles.