Map completion from partial observation using the global structure of multiple environmental maps

TL;DR

This paper introduces MCN-SLAM, a novel SLAM approach that leverages deep neural networks and generative adversarial networks to utilize the global structure of maps, significantly improving map estimation from partial observations.

Contribution

The paper presents a new SLAM method that incorporates deep generative models for map completion, enhancing efficiency and accuracy over traditional approaches.

Findings

Estimates environment maps 1.3 times better than previous SLAM methods.

Utilizes GAN-based map completion networks trained on large map datasets.

Effectively leverages global environmental structure for improved map reconstruction.

Abstract

Using the spatial structure of various indoor environments as prior knowledge, the robot would construct the map more efficiently. Autonomous mobile robots generally apply simultaneous localization and mapping (SLAM) methods to understand the reachable area in newly visited environments. However, conventional mapping approaches are limited by only considering sensor observation and control signals to estimate the current environment map. This paper proposes a novel SLAM method, map completion network-based SLAM (MCN-SLAM), based on a probabilistic generative model incorporating deep neural networks for map completion. These map completion networks are primarily trained in the framework of generative adversarial networks (GANs) to extract the global structure of large amounts of existing map data. We show in experiments that the proposed method can estimate the environment map 1.3 times better than the previous SLAM methods in the situation of partial observation.