Online Domain Adaptation for Occupancy Mapping

TL;DR

This paper introduces a novel framework based on optimal transport theory for real-time domain adaptation of occupancy maps, reducing training costs and enabling scalable autonomous navigation in diverse urban environments.

Contribution

It proposes a theoretical approach for adapting occupancy map parameters across environments, significantly lowering computational costs compared to traditional methods.

Findings

Parameters can be adapted with negligible computational cost.

Framework effectively transfers knowledge from simulation to real-world data.

Enables large-scale probabilistic mapping in urban environments.

Abstract

Creating accurate spatial representations that take into account uncertainty is critical for autonomous robots to safely navigate in unstructured environments. Although recent LIDAR based mapping techniques can produce robust occupancy maps, learning the parameters of such models demand considerable computational time, discouraging them from being used in real-time and large-scale applications such as autonomous driving. Recognizing the fact that real-world structures exhibit similar geometric features across a variety of urban environments, in this paper, we argue that it is redundant to learn all geometry dependent parameters from scratch. Instead, we propose a theoretical framework building upon the theory of optimal transport to adapt model parameters to account for changes in the environment, significantly amortizing the training cost. Further, with the use of high-fidelity driving simulators and real-world datasets, we demonstrate how parameters of 2D and 3D occupancy maps can be automatically adapted to accord with local spatial changes. We validate various domain adaptation paradigms through a series of experiments, ranging from inter-domain feature transfer to simulation-to-real-world feature transfer. Experiments verified the possibility of estimating parameters with a negligible computational and memory cost, enabling large-scale probabilistic mapping in urban environments.