scholarly journals Loop Closure Detection in RGB-D SLAM by Utilizing Siamese ConvNet Features

2021 ◽  
Vol 12 (1) ◽  
pp. 62
Author(s):  
Gang Xu ◽  
Xiang Li ◽  
Xingyu Zhang ◽  
Guangxin Xing ◽  
Feng Pan
Keyword(s):  
Detection Methods ◽  
Position Information ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Depth Images ◽  
Localization And Mapping ◽  
L2 Norm ◽  
Significant Research ◽  
Key Issues ◽  
Public Datasets

Loop closure detection is a key challenge in visual simultaneous localization and mapping (SLAM) systems, which has attracted significant research interest in recent years. It entails correctly determining whether a scene has previously been visited by a mobile robot and completely establishing the consistent maps of motion. There are many loop closure detection methods that have been proposed, but most of these algorithms are handcrafted features-based and perform weak robustness to illumination variations. In this paper, we investigate a Siamese Convolutional Neural Network (SCNN) to solve the task of loop closure detection in RGB-D SLAM. Firstly, we use a pre-trained SCNN model to extract features as image descriptors; then, the L2 norm distance is adopted as a similarity metric between descriptors. In terms of the learned features for matching, there are two key issues for discussion: (1) how to define an appropriate loss as supervision (utilizing the cross-entropy loss, the contrastive loss, or the combination of two); and (2) how to combine the appearance information in RGB images and position information in depth images (utilizing early fusion, mid-level fusion or late fusion). We compare our proposed method of different baseline by experiments carried out on two public datasets (New College and NYU), and our performance outperforms the state-of-the-art.

scholarly journals SVG-Loop: Semantic–Visual–Geometric Information-Based Loop Closure Detection

2021 ◽  
Vol 13 (17) ◽  
pp. 3520
Author(s):  
Zhian Yuan ◽  
Ke Xu ◽  
Xiaoyu Zhou ◽  
Bin Deng ◽  
Yanxin Ma
Keyword(s):  
Detection Algorithm ◽  
Detection Methods ◽  
Vector Model ◽  
Limited Information ◽  
Complex Environments ◽  
Dynamic Features ◽  
Geometric Information ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Localization And Mapping

Loop closure detection is an important component of visual simultaneous localization and mapping (SLAM). However, most existing loop closure detection methods are vulnerable to complex environments and use limited information from images. As higher-level image information and multi-information fusion can improve the robustness of place recognition, a semantic–visual–geometric information-based loop closure detection algorithm (SVG-Loop) is proposed in this paper. In detail, to reduce the interference of dynamic features, a semantic bag-of-words model was firstly constructed by connecting visual features with semantic labels. Secondly, in order to improve detection robustness in different scenes, a semantic landmark vector model was designed by encoding the geometric relationship of the semantic graph. Finally, semantic, visual, and geometric information was integrated by fuse calculation of the two modules. Compared with art-of-the-state methods, experiments on the TUM RBG-D dataset, KITTI odometry dataset, and practical environment show that SVG-Loop has advantages in complex environments with varying light, changeable weather, and dynamic interference.

scholarly journals Incremental Loop Closure Verification by Guided Sampling

2017 ◽  
Vol 21 (1) ◽  
pp. 59-66
Author(s):  
Tanaka Kanji ◽  
Keyword(s):  
Image Retrieval ◽  
Good Precision ◽  
Sampling Strategies ◽  
Detection Problem ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Robot Trajectory ◽  
Localization And Mapping ◽  
Random Sample Consensus ◽  
Closure Constraints

Loop closure detection, which is the task of identifying locations revisited by a robot in a sequence of odometry and perceptual observations, is typically formulated as a combination of two subtasks: (1) bag-of-words image retrieval and (2) post-verification using random sample consensus (RANSAC) geometric verification. The main contribution of this study is the proposal of a novel post-verification framework that achieves good precision recall trade-off in loop closure detection. This study is motivated by the fact that not all loop closure hypotheses are equally plausible (e.g., owing to mutual consistency between loop closure constraints) and that if we have evidence that one hypothesis is more plausible than the others, then it should be verified more frequently. We demonstrate that the loop closure detection problem can be viewed as an instance of a multi-model hypothesize-and-verify framework. Thus, we can build guided sampling strategies on this framework where loop closures proposed using image retrieval are verified in a planned order (rather than in a conventional uniform order) to operate in a constant time. Experimental results using a stereo simultaneous localization and mapping (SLAM) system confirm that the proposed strategy, the use of loop closure constraints and robot trajectory hypotheses as a guide, achieves promising results despite the fact that there exists a significant number of false positive constraints and hypotheses.

scholarly journals Multi-Objective Optimization of Loop Closure Detection Parameters for Indoor 2D Simultaneous Localization and Mapping

Sensors ◽  
2020 ◽  
Vol 20 (7) ◽  
pp. 1906
Author(s):  
Dongxiao Han ◽  
Yuwen Li ◽  
Tao Song ◽  
Zhenyang Liu
Keyword(s):  
Simultaneous Localization And Mapping ◽  
Ground Truth ◽  
Optimization Method ◽  
Evaluation Metrics ◽  
Multi Objective Optimization ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Multi Objective ◽  
Evaluation Approach ◽  
Localization And Mapping

Aiming at addressing the issues related to the tuning of loop closure detection parameters for indoor 2D graph-based simultaneous localization and mapping (SLAM), this article proposes a multi-objective optimization method for these parameters. The proposed method unifies the Karto SLAM algorithm, an efficient evaluation approach for map quality with three quantitative metrics, and a multi-objective optimization algorithm. More particularly, the evaluation metrics, i.e., the proportion of occupied grids, the number of corners and the amount of enclosed areas, can reflect the errors such as overlaps, blurring and misalignment when mapping nested loops, even in the absence of ground truth. The proposed method has been implemented and validated by testing on four datasets and two real-world environments. For all these tests, the map quality can be improved using the proposed method. Only loop closure detection parameters have been considered in this article, but the proposed evaluation metrics and optimization method have potential applications in the automatic tuning of other SLAM parameters to improve the map quality.

scholarly journals A Light Visual Mapping and Navigation Framework for Low-Cost Robots

2015 ◽  
Vol 24 (4) ◽  
pp. 505-524 ◽  
Author(s):  
Stephane Bazeille ◽  
Emmanuel Battesti ◽  
David Filliat
Keyword(s):  
Large Range ◽  
Low Cost ◽  
Indoor Environments ◽  
Noise Levels ◽  
Visual Sensor ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Localization And Mapping ◽  
Computational Resources

AbstractWe address the problems of localization, mapping, and guidance for robots with limited computational resources by combining vision with the metrical information given by the robot odometry. We propose in this article a novel light and robust topometric simultaneous localization and mapping framework using appearance-based visual loop-closure detection enhanced with the odometry. The main advantage of this combination is that the odometry makes the loop-closure detection more accurate and reactive, while the loop-closure detection enables the long-term use of odometry for guidance by correcting the drift. The guidance approach is based on qualitative localization using vision and odometry, and is robust to visual sensor occlusions or changes in the scene. The resulting framework is incremental, real-time, and based on cheap sensors provided on many robots (a camera and odometry encoders). This approach is, moreover, particularly well suited for low-power robots as it is not dependent on the image processing frequency and latency, and thus it can be applied using remote processing. The algorithm has been validated on a Pioneer P3DX mobile robot in indoor environments, and its robustness is demonstrated experimentally for a large range of odometry noise levels.

scholarly journals Role of Deep Learning in Loop Closure Detection for Visual and Lidar SLAM: A Survey

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1243
Author(s):  
Saba Arshad ◽  
Gon-Woo Kim
Keyword(s):  
Deep Learning ◽  
Detailed Comparison ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Detection Algorithms ◽  
Mobile Objects ◽  
Localization And Mapping ◽  
Loop Detection ◽  
Scene Representation ◽  
Matching Strategy

Loop closure detection is of vital importance in the process of simultaneous localization and mapping (SLAM), as it helps to reduce the cumulative error of the robot’s estimated pose and generate a consistent global map. Many variations of this problem have been considered in the past and the existing methods differ in the acquisition approach of query and reference views, the choice of scene representation, and associated matching strategy. Contributions of this survey are many-fold. It provides a thorough study of existing literature on loop closure detection algorithms for visual and Lidar SLAM and discusses their insight along with their limitations. It presents a taxonomy of state-of-the-art deep learning-based loop detection algorithms with detailed comparison metrics. Also, the major challenges of conventional approaches are identified. Based on those challenges, deep learning-based methods were reviewed where the identified challenges are tackled focusing on the methods providing long-term autonomy in various conditions such as changing weather, light, seasons, viewpoint, and occlusion due to the presence of mobile objects. Furthermore, open challenges and future directions were also discussed.

scholarly journals A LiDAR/Visual SLAM Backend with Loop Closure Detection and Graph Optimization

2021 ◽  
Vol 13 (14) ◽  
pp. 2720
Author(s):  
Shoubin Chen ◽  
Baoding Zhou ◽  
Changhui Jiang ◽  
Weixing Xue ◽  
Qingquan Li
Keyword(s):  
Point Clouds ◽  
Visual Features ◽  
Visual Slam ◽  
Key Factors ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Active Sensor ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
Pose Graph Optimization

LiDAR (light detection and ranging), as an active sensor, is investigated in the simultaneous localization and mapping (SLAM) system. Typically, a LiDAR SLAM system consists of front-end odometry and back-end optimization modules. Loop closure detection and pose graph optimization are the key factors determining the performance of the LiDAR SLAM system. However, the LiDAR works at a single wavelength (905 nm), and few textures or visual features are extracted, which restricts the performance of point clouds matching based loop closure detection and graph optimization. With the aim of improving LiDAR SLAM performance, in this paper, we proposed a LiDAR and visual SLAM backend, which utilizes LiDAR geometry features and visual features to accomplish loop closure detection. Firstly, the bag of word (BoW) model, describing the visual similarities, was constructed to assist in the loop closure detection and, secondly, point clouds re-matching was conducted to verify the loop closure detection and accomplish graph optimization. Experiments with different datasets were carried out for assessing the proposed method, and the results demonstrated that the inclusion of the visual features effectively helped with the loop closure detection and improved LiDAR SLAM performance. In addition, the source code, which is open source, is available for download once you contact the corresponding author.

Online appearance-based place recognition and mapping algorithms for autonomous robot navigation

2021 ◽  
Author(s):  
Κωνσταντίνος Τσιντώτας
Keyword(s):  
Image Processing ◽  
Robot Navigation ◽  
Simultaneous Localization And Mapping ◽  
Autonomous Robot ◽  
Place Recognition ◽  
Mapping Algorithms ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Localization And Mapping ◽  
Autonomous Robot Navigation

Τα αυτόνομα ρομποτικά συστήματα αποτελούν ένα από τα τμήματα που βρίσκονται στην αιχμή της τεχνολογίας και της έρευνας. Τα σύγχρονα ρομποτικά συστήματα που δημιουργούνται δύναται να έχουν την ικανότητα να μπορούν να προσδιορίσουν την θέση τους μέσα στο περιβάλλον το οποίο περιπλανιούνται. Η επίτευξη του συγκεκριμένου στόχου επιτυγχάνεται μέσω της κατασκευής ενός χάρτη την στιγμή που πλοηγείται, μέσω ενός μηχανισμού που είναι ευρέως γνωστός ως ταυτόχρονος προσδιορισμός θέσης και τοποθεσίας (Simultaneous Localization and Mapping –SLAM). Ο χάρτης δημιουργείται μέσω των μετρήσεων που δέχεται το σύστημα από τους επιμέρους αισθητήρες που είναι τοποθετημένοι επάνω στο ρομποτικό σύστημα. Η αύξηση της υπολογιστικής ισχύς τα τελευταία χρόνια και η ευρέως διαδεδομένη χρήση των καμερών οδήγησε στα αυτόνομα συστήματα να επιλέγεται ως κύρια πηγή απόκτησης δεδομένων η χρήση καμερών. Καθώς όμως το ρομπότ διασχίζει την τροχιά του μέσα σε ένα άγνωστο περιβάλλον είναι έντονος ο κίνδυνος δημιουργίας ενός χάρτη με εσφαλμένα στοιχεία στην σχεδιασμένη τροχιά, με κύρια αίτια που μπορεί να σχετίζονται στην κακή εκτίμηση μετρήσεων των αισθητήριων ή σε ενδεχόμενη δυσλειτουργία των ενσωματωμένων οργάνων του συστήματος. Το πρόβλημα που αναφέρθηκε αποτελεί κίνδυνο για την περάτωση της αποστολής του ρομποτικού συστήματος, το οποίο όμως μπορεί να διορθωθεί με επαναπροσδιορισμό της τροχιάς του εφόσον είναι εφικτή η αναγνώριση της περιοχής που διασχίζει μέσω των οπτικών αισθητηρίων που διαθέτει. Τα συστήματα τα οποία βασίζονται αποκλειστικά στις κάμερες ως μέσα αναγνώρισης περιοχής είναι γνωστά ως εμφάνισης βασιζόμενα συστήματα (appearance based systems) και οι τεχνικές που προσπαθούν να λύσουν το πρόβλημα είναι γνωστές ως ανίχνευση κλειστών βρόγχων (loop closure detection). Η παρούσα διδακτορική διατριβή εστιάζει στην δημιουργία ενός αντίστοιχου συστήματος ικανού να αναγνωρίσει μια επαναλαμβανόμενα επισκεπτόμενη περιοχή μέσω των εισερχόμενων εικόνων που δέχεται. Η επίτευξη του συστήματος στηρίζεται σε τρείς σημαντικές μονάδες. Στην αρχή έρχεται η μονάδα επεξεργασίας εικόνας (image processing) με την οποία είναι εφικτή η ερμηνεία των εισερχόμενων εικόνων. Μέσω τεχνικών περιγραφής εικόνας είναι δυνατή η λήψη σημείων ενδιαφέροντος ικανά να περιγράψουν την σκηνή που αντιμετωπίζει το ρομπότ. Κάθε εικόνα ενδέχεται να περιέχει εκατοντάδες τοπικά σημεία ενδιαφέροντος με αποτέλεσμα να μπορεί να περιγραφεί ως ένα σύνολο από επιμέρους χαρακτηριστικά. Ο χάρτης (map) αποτελεί την μονάδα που είναι υπεύθυνη για την αναπαράσταση του κόσμου που γνωρίζει το ρομπότ με βάση τα χαρακτηριστικά που έχουν ληφθεί από τις σκηνές που έχει επισκεφθεί. Τέλος, έρχεται η μονάδα πεποίθησης πιθανής περιοχής (belief generator), η οποία συνδυάζει τα εισερχόμενα δεδομένα με τα δεδομένα του χάρτη ώστε να είναι σε θέση να λάβει την απόφαση για το αν το σύστημα βρίσκεται ή όχι σε οικεία περιοχή με σκοπό να επαναπροσδιορίσει την τροχιά του.

Robust Loop Closure Detection Using Bayes Filters and CNN Features

2019 ◽  
pp. 1950011
Author(s):  
Qiang Liu ◽  
Fuhai Duan
Keyword(s):  
Feature Fusion ◽  
Sequence Matching ◽  
Severe Condition ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Localization And Mapping ◽  
Penalty Factor ◽  
Sequential Information ◽  
Bidirectional Propagation ◽  
Different Characteristics

This paper focuses on loop-closure detection (LCD) for a visual simultaneous localization and mapping (SLAM) system. We present a strategy that combines a Bayes filter and features from a pre-trained convolution neural network (CNN) to perform LCD. Rather than using features from only one layer, we fuse features from multiple layers based on spatial pyramid pooling. A flexible Bayes model is then formulated to integrate the sequential information and similarities that are computed by features at different scales. The introduction of a penalty factor and bidirectional propagation enables our approach to handle complex trajectories. We present extensive experiments on challenging datasets, and we compare our approach to state-of-the-art methods, to evaluate it. The results show that our approach can ensure remarkable performance under severe condition changes and handle trajectories that have different characteristics. We also show the advantages of Bayes filters over sequence matching in the experiments, and we analyze our feature fusion strategy by visualizing the activations of the CNN.

scholarly journals Robust Loop Closure Detection Integrating Visual–Spatial–Semantic Information via Topological Graphs and CNN Features

2020 ◽  
Vol 12 (23) ◽  
pp. 3890
Author(s):  
Yuwei Wang ◽  
Yuanying Qiu ◽  
Peitao Cheng ◽  
Xuechao Duan
Keyword(s):  
Semantic Information ◽  
Graph Matching ◽  
Spatial Relationships ◽  
Dynamic Scenes ◽  
Topological Graphs ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Visual Spatial ◽  
Localization And Mapping ◽  
The Impact

Loop closure detection is a key module for visual simultaneous localization and mapping (SLAM). Most previous methods for this module have not made full use of the information provided by images, i.e., they have only used the visual appearance or have only considered the spatial relationships of landmarks; the visual, spatial and semantic information have not been fully integrated. In this paper, a robust loop closure detection approach integrating visual–spatial–semantic information is proposed by employing topological graphs and convolutional neural network (CNN) features. Firstly, to reduce mismatches under different viewpoints, semantic topological graphs are introduced to encode the spatial relationships of landmarks, and random walk descriptors are employed to characterize the topological graphs for graph matching. Secondly, dynamic landmarks are eliminated by using semantic information, and distinctive landmarks are selected for loop closure detection, thus alleviating the impact of dynamic scenes. Finally, to ease the effect of appearance changes, the appearance-invariant descriptor of the landmark region is extracted by a pre-trained CNN without the specially designed manual features. The proposed approach weakens the influence of viewpoint changes and dynamic scenes, and extensive experiments conducted on open datasets and a mobile robot demonstrated that the proposed method has more satisfactory performance compared to state-of-the-art methods.

Sign in / Sign up

Export Citation Format

Share Document