ORBSLAM-Atlas: a robust and accurate multi-map system

Abstract

We propose ORBSLAM-Atlas, a system able to handle an unlimited number of disconnected sub-maps, that includes a robust map merging algorithm able to detect submaps with common regions and seamlessly fuse them. The outstanding robustness and accuracy of ORBSLAM are due to its ability to detect wide-baseline matches between keyframes, and to exploit them by means of non-linear optimization, however it only can handle a single map. ORBSLAM-Atlas brings the wide-baseline matching detection and exploitation to the multiple map arena. The result is a SLAM system significantly more general and robust, able to perform multisession mapping. If tracking is lost during exploration, instead of freezing the map, a new sub-map is launched, and it can be fused with the previous map when common parts are visited. Our criteria to declare the camera lost contrast with previous approaches that simply count the number of tracked points, we propose to discard also inaccurately estimated camera poses due to bad geometrical conditioning. As a result, the map is split into more accurate sub-maps, that are eventually merged in a more accurate global map, thanks to the multi-mapping capabilities.We provide extensive experimental validation in the EuRoC datasets, where ORBSLAM-Atlas obtains accurate monocular and stereo results in the difficult sequences where ORBSLAM failed. We also build global maps after multiple sessions in the same room, obtaining the best results to date, between 2 and 3 times more accurate than competing multi-map approaches. We also show the robustness and capability of our system to deal with dynamic scenes, quantitatively in the EuRoC datasets and qualitatively in a densely populated corridor where camera occlusions and tracking losses are frequent.We provide a quantitative experimental validation in the EuRoC datasets, in which ORBSLAM-Atlas achieves the best results to date for a global map after multiple sessions. In the monocular EuRoC difficult datasets, it greatly improves the coverage and localization error when compared with the single map ORBSLAM. Additionally, the system has proved outstanding robustness in dealing with dynamic scenes.

DGS-SLAM: A Fast and Robust RGBD SLAM in Dynamic Environments Combined by Geometric and Semantic Information

“…ORB-SLAM2 improves and optimizes feature point extraction and keyframe selection, which has good operation efficiency and stability. Based on ORB-SLAM2, ORB-SLAM3 [35] tightly integrates visual and inertial information and adds a multiple map system (ATLAS [36]), which can work in real-time in various environments. However, these methods cannot distinguish between static and dynamic targets in the scene, which can cause VSLAM systems to degrade the accuracy of localization and map building due to incorrect data association.…”

RDS-SLAM: Real-Time Dynamic SLAM Using Semantic Segmentation Methods

“…This system estimates the ego-motion of the camera by matching the corresponding ORB [17] features between the current frame and previous frames and has three parallel threads: tracking, local mapping, and loop closing. Carlos et al proposed the latest version ORB-SLAM3 [18], mainly adding two novelties: 1) a feature-based tightly-integrated visual-inertial SLAM that fully relies on maximum-a-posteriori (MAP) estimation; 2) a multiple map system (ATLAS [19]) that relies on a new place recognition method with improved recall. Apart from featurebased methods, many direct vSLAM solutions [20]- [23] are also proposed.…”

“…The Atlas [19] is a multi-map representation that handles an unlimited number of sub-maps. Two kinds of maps, active map and non-active map, are managed in the atlas.…”

RDMO-SLAM: Real-Time Visual SLAM for Dynamic Environments Using Semantic Label Prediction With Optical Flow

“…Based on ORB-SLAM, ORB-SLAM2 [22], a complete SLAM system for monocular, stereo, and RGB-D camera was presented, which can work in real-time in various environments. Carlos et al proposed the latest version of ORB-SLAM, ORB-SLAM3 [23], which tightly integrates visual and inertial information and adds a multiple map system (ATLAS [24]). Our previous work, a real-time dynamic SLAM using semantic segmentation methods (RDS-SLAM) [18], is implemented based on ORB-SLAM3.…”

“…Our previous work, RDS-SLAM [18] implemented based on ORB-SLAM3, adds a novel semantic tracking thread that segments objects with the aid of Mask R-CNN or SegNet, and then uses the semantic information to update and propagate the moving probability of map points in ATLAS [24]. We follow the ideas of RDS-SLAM and mainly solve the insufficient semantic information problem when using Mask R-CNN.…”