Search

Loop Closure (LC) is a crucial task in Simultaneous Localization and Mapping (SLAM) for Autonomous Ground Vehicles (AGV). It is an active research area because it improves global localization efficiency. The consistency of the global map and the accuracy of the AGV's location in an unknown environment are highly correlated with the efficiency and robustness of Loop Closure Detection (LCD), especially when facing environmental changes or data unavailability. We propose to introduce multimodal complementary data to increase the algorithms' resilience. Various methods using different data sources have been proposed to achieve precise place recognition. However, integrating a multimodal loop-closure fusion process that combines multiple information sources within a SLAM system has been explored less. Additionally, existing multimodal place recognition techniques are often difficult to integrate into existing frameworks. In this paper, we propose a fusion scheme of multiple place recognition methods based on camera and LiDAR data for a robust multimodal LCD. The presented approach uses Similarity-Guided Particle Filtering (SGPF) to identify and verify candidates for loop closure. Based on the ORB-SLAM2 framework, the proposed method uses two perception sensors (camera and LiDAR) under two data representation models for each. Our experiments on both KITTI and a self-collected dataset show that our approach outperforms the state-of-the-art methods in terms of place recognition metrics or localization accuracy metrics. The proposed Multi-Modal Loop Closure (MMLC) framework enhances the robustness and accuracy of AGV's localization by fusing multiple sensor modalities, ensuring consistent performance across diverse environments. Its real-time operation and early loop closure detection enable timely trajectory corrections, reducing navigation errors and supporting cost-effective deployment with adaptable sensor configurations. [ABSTRACT FROM AUTHOR]

Loop closure detection, also known as place recognition, a key component of simultaneous localization and mapping (SLAM) systems, aims to recognize previously visited locations and reduce the accumulated drift error caused by odometry. Current vision-based methods are susceptible to variations in illumination and perspective, limiting their generalization ability and robustness. Thus, in this paper, we propose CrossT-Net (Cross Transformer Net), a novel cross-attention based loop closure detection network for LiDAR. CrossT-Net directly estimates the similarity between two frames by leveraging multi-class information maps, including range, intensity, and normal maps, to comprehensively characterize environmental features. A Siamese Encoder Net with shared parameters extracts frame features, and a Cross Transformer module captures intra-frame context and inter-frame correlations through self-attention and cross-attention mechanisms. In the final stage, an Overlap Estimation Module predicts the point cloud overlap between two frames. Experimental results on several benchmark datasets demonstrate that our proposed method outperforms existing methods in precision and recall, and exhibits strong generalization performance in different road environments. The implementation of our approach is available at: . [ABSTRACT FROM AUTHOR]

Efficiently reconstructing complex and intricate surfaces at scale remains a significant challenge in 3D surface reconstruction. Recently, implicit neural representations have become a popular topic in 3D surface reconstruction. However, how to handle loop closure and bundle adjustment is a tricky problem for neural methods, because they learn the neural parameters globally. We present an algorithm that leverages the concept of surfels and expands relevant definitions to address such challenges. By integrating neural descriptors with surfels and framing surfel association as a deformation graph optimization problem, our method is able to effectively perform loop closure detection and loop correction in challenging scenarios. Furthermore, the surfel-level representation simplifies the complexity of 3D neural reconstruction. Meanwhile, the binding of neural descriptors to corresponding surfels produces a dense volumetric signed distance function (SDF), enabling the mesh reconstruction. Our approach demonstrates a significant improvement in reconstruction accuracy, reducing the average error by 16.9% compared to previous methods, while also generating modeling files that are up to 90% smaller than those produced by traditional methods. [ABSTRACT FROM AUTHOR]

Deep learning models centered on retrieval have made significant strides in point cloud place recognition. However, existing approaches struggle to generate discriminative global descriptors and often rely on labor-intensive negative sample mining. Such constraints limit their usability in dynamic and open-world scenarios. To address these challenges, we introduce LAWS, a pioneering classification-centric neural framework that emphasizes looking at the whole scene for superior point cloud descriptor extraction. Central to our approach is the space partitioning design, constructed to provide holistic scene supervision, ensuring the comprehensive learning of scene features. To counteract potential ambiguities arising from the single orthogonal partition boundary, a complementary mechanism of repartitioning space diagonally is specifically designed to dispel classification uncertainties. Under the enhanced partitioning mechanism, the space is separated into several classes and groups. Furthermore, to prevent knowledge forgetting between different groups, a special training strategy is employed, allowing for distinct training of each group. The extensive experiments, encompassing both indoor and outdoor settings and different tasks, validate the generality of LAWS. It not only outperforms contemporary methods but also demonstrates a profound generalization ability across various unseen environments and sensor modalities. Our method achieves a 2.6% higher average top-1 recall on Oxford RobotCar Dataset and a 7.8% higher average recall when generalized to In-house Dataset compared with retrieval-based methods. Furthermore, LAWS also outperforms retrieval-based methods in terms of F 1 score, with improvements of 12.7 and 29.2 on the MulRan and KITTI datasets, respectively. Notably, the average localization accuracy of LAWS in indoor environments reached about 68.1%. Moreover, the scalability and efficiency places LAWS in a leading position for continuous exploration and long-term autonomy. Our code is available at https://github.com/BrusonX/LAWS. [ABSTRACT FROM AUTHOR]

Indoor environments are considered challenging due to their perceived degradation. This is attributed to both the homogeneity of indoor scene structures and the scarcity of feature points, particularly in long corridor environments. Additionally, low-resolution 3D laser scanners may not effectively capture the real information of the indoor environment. To address these issues, this paper proposes a tightly coupled SLAM method based on factor graph optimization for laser-inertial fusion. Firstly, a dynamic feature point extraction method is designed to adaptively adjust the feature extraction quantity by detecting indoor degradation, constructing rich and robust feature constraints to enhance pose estimation accuracy. Next, robust and precise SLAM in indoor environments is achieved through factor graph optimization. Finally, validation is conducted on the KITTI dataset and in real scenarios. Experimental results demonstrate that the proposed method exhibits superior accuracy, real-time performance, and robustness compared to ALOAM, LeGO-LOAM, FLOAM, and LIO-SAM in terms of localization, mapping precision, and global consistency. The method effectively reduces cumulative errors and ensures the global consistency of map construction.

This paper presents open-source software for large-scale 3D mapping using an open-hardware hand-held measurement device. This work is dedicated to educational and research purposes. This software is composed of three components: LIDAR odometry, single-session refinement and multi-session refinement. Data refinement uses a pose-graph loop closure technique and an Iterative Closest Point algorithm to minimize the error of the edge. The results are 3D point clouds in LAZ data format (compressed LAS - LIDAR Aerial Survey). It was tested in many real-world scenarios/applications: city-level 3D mapping, culture heritage, creating ground truth data for mobile robots, precise forestry, and large-scale indoor 3D mapping. This software can run on Linux and Windows machines, it does not incorporate GPU computing. It is advised to use at least 32 GB of RAM to cope with large data sets. The optimization framework is implemented from scratch using the Eigen library, thus there is not need to install any additional libraries such as Ceres, g2o, GTSAM, manif, Sophus etc.

Aiming at the problem that the VINS-MONO (monoculor visual-inertial state) algorithm has low positioning accuracy and the sparse point cloud map constructed contains less information and cannot be used for autonomous navigation of unmanned aerial vehicles, a centralized multi-UAV (unmanned aerial vehicle) collaborative positioning and dense map construction algorithm was proposed. In order to establish the connection constraint between multiple UAVs, the algorithm used four different types of coordinate systems for coordinate transformation optimization, and used the two-way detection matching method to match the feature points of key frames, combined with the PROSAC (progressive sampling consensus) algorithm to eliminate false matching. By optimizing the global pose map of multiple UAVs with loopback closure constraints, a global dense map for collaborative navigation of five UAVs was constructed by combining Voxblox. In the five sequences of the EuRoc dataset, the proposed multi-UAV co-localization and dense map construction algorithm reduced the absolute trajectory error by 34%, 26%, 32%, 24% and 19% compared with the VINS-MONO algorithm, respectively. It is proved by experiments that the improved algorithm effectively improves the positioning accuracy between multiple UAVs, and the constructed globally consistent dense map contains distance information and gradient information, which can be used for autonomous navigation of multiple UAVs. [ABSTRACT FROM AUTHOR]

In this paper, a visual simultaneous localization and mapping (VSLAM/visual SLAM) system called underwater visual SLAM (UVS) system is presented, specifically tailored for camera-only navigation in natural underwater environments. The UVS system is particularly optimized towards precision and robustness, as well as lifelong operations. We build upon Oriented features from accelerated segment test and Rotated Binary robust independent elementary features simultaneous localization and mapping (ORB-SLAM) and improve the accuracy by performing an exact search in the descriptor space during triangulation and the robustness by utilizing a unified initialization method and a motion model. In addition, we present a scale-agnostic station-keeping detection, which aims to optimize the map and poses during station-keeping, and a pruning strategy, which takes into account the point's age and distance to the active keyframe. An exhaustive evaluation is presented to the reader, using a total of 38 in-air and underwater sequences. [ABSTRACT FROM AUTHOR]

We analyse, for the first time, the popular loop closing module of a well known and widely used open-source visual SLAM (ORB-SLAM) pipeline. Investigating failures in the loop closure module of visual SLAM is challenging since it consists of multiple building blocks. Our meticulous investigations have revealed a few interesting findings. Contrary to reported results, ORB-SLAM frequently misses large fraction of loop closures on public (KITTI, TUM RGB-D) datasets. One common assumption is, in such scenarios, the visual place recognition (vPR) block of the loop closure module is unable to find a suitable match due to extreme conditions (dynamic scene, viewpoint/scale changes). We report that native vPR of ORB-SLAM is not the sole reason for these failures. Although recent deep vPR alternatives achieve impressive matching performance, replacing native vPR with these deep alternatives will only partially improve loop closure performance of visual SLAM. Our findings suggest that the problem lies with the subsequent relative pose estimation module between the matching pair. ORB-SLAM3 has improved the recall of the original loop closing module. However, even in ORB-SLAM3, the loop closing module is the major reason behind loop closing failures. Surprisingly, using off-the-shelf ORB and SIFT based relative pose estimators (non real-time) manages to close most of the loops missed by ORB-SLAM. This significant performance gap between the two available methods suggests that ORB-SLAM's pipeline can be further matured by focusing on the relative pose estimators, to improve loop closure performance, rather than investing more resources on improving vPR. We also evaluate deep alternatives for relative pose estimation in the context of loop closures. Interestingly, the performance of deep relocalization methods (e.g. MapNet) is worse than classic methods even in loop closures scenarios. This finding further supports the fundamental limitation of deep relocalization methods recently diagnosed. Finally, we expose bias in well-known public dataset (KITTI) due to which these commonly occurring failures have eluded the community. We augment the KITTI dataset with detailed loop closing labels. In order to compensate for the bias in the public datasets, we provide a challenging loop closure dataset which contains challenging yet commonly occurring indoor navigation scenarios with loop closures. We hope our findings and the accompanying dataset will help the community in further improving the popular ORB-SLAM's pipeline. [ABSTRACT FROM AUTHOR]

Some structures in the harbour environment need to be inspected regularly. However, these scenarios present a major challenge for the accurate estimation of a vehicle's position and subsequent recognition of similar images. In these scenarios, visibility can be poor, making place recognition a difficult task as the visual appearance of a local feature can be compromised. Under these operating conditions, imaging sonars are a promising solution. The quality of the captured images is affected by some factors but they do not suffer from haze, which is an advantage. Therefore, a purely acoustic approach for unsupervised recognition of similar images based on forward-looking sonar (FLS) data is proposed to solve the perception problems in harbour facilities. To simplify the variation of environment parameters and sensor configurations, and given the need for online data for these applications, a harbour scenario was recreated using the Stonefish simulator. Therefore, experiments were conducted with preconfigured user trajectories to simulate inspections in the vicinity of structures. The place recognition approach performs better than the results obtained from optical images. The proposed method provides a good compromise in terms of distinctiveness, achieving 87.5% recall considering appropriate constraints and assumptions for this task given its impact on navigation success. That is, it is based on a similarity threshold of 0.3 and 12 consistent features to consider only effective loops. The behaviour of FLS is the same regardless of the environment conditions and thus this work opens new horizons for the use of these sensors as a great aid for underwater perception, namely, to avoid degradation of navigation performance in muddy conditions. [ABSTRACT FROM AUTHOR]

The European Commission (EC) originally proposed ideas for a European Galileo satellite navigation system in 1999. A four-phase development is planned, involving investment from both the public and commercial sectors. Galileo is intended for both public and government use; the system is administered and operated by civil administration. Galileo will consist of a constellation of 30 satellites, a number of globally situated ground stations, and a ground control and monitoring system - all of which are quite similar to the structure, format, and layout of GPS. This document discusses an experiment at the project site that used the static approach to integrate GPS, GLONASS, GALILEO, Beidou, and QZSS signals. This research analyses the possible precision of GPS-only and GPS/GLONASS/GALILEO/Beidou/QZSS. These results suggest that combining a GPS system with GALILEO, GLONASS Beidou, and QZSS is preferable for surveying purposes. Integrating GPS/GLONASS/GALILEO/Beidou/QZSS static measurements in the study region with 0-120 millimetre accuracy looks to be possible in three days. [ABSTRACT FROM AUTHOR]

For autonomous vehicles, real-time localization and mapping in the unknown environment is very important. In this paper, a fast and lightweight 3D LiDAR simultaneously localization and mapping (SLAM) is presented for the localization of autonomous vehicles in large-scale urban environments. A novel encoding approach based on depth information is proposed to encode unordered point clouds with various resolutions, which avoids missing dimensional information in the projection of point clouds onto a 2D plane. Principal components analysis (PCA) is modified by dynamically selecting neighborhood points according to the encoded depth information to fit the local plane with less time consuming. The threshold and the number of feature points are adaptive according to distance intervals, results in sparse feature points extracted and uniformly distributed in the three-dimensional space. The extracted significant feature points improve the accuracy of the odometer and speed up the alignment of the point cloud. The effectiveness and robustness of the proposed algorithm are verified on the KITTI odometry benchmark and MVSECD. A fast runtime of 21 ms is obtained for the odometer estimation. Compared to several typical state-of-the-art methods on the KITTI odometry benchmark, the proposed approach reduces translation errors by at least 19% and rotation errors by 7.1%.

Designing movesets providing high quality protein conformations remains a hard problem, especially when it comes to deform a long protein backbone segment, and a key building block to do so is the so‐called tripeptide loop closure (TLC). Consider a tripeptide whose first and last bonds (N1Cα;1 and Cα;3C3) are fixed, and so are all internal coordinates except the six ϕψi=1,2,3 dihedral angles associated to the three Cα carbons. Under these conditions, the TLC algorithm provides all possible values for these six dihedral angles–there exists at most 16 solutions. TLC moves atoms up to ∼5Å in one step and retains low energy conformations, whence its pivotal role to design move sets sampling protein loop conformations. In this work, we relax the previous constraints, allowing the last bond (Cα;3C3) to freely move in 3D space–or equivalently in a 5D configuration space. We exhibit necessary geometric constraints in this 5D space for TLC to admit solutions. Our analysis provides key insights on the geometry of solutions for TLC. Most importantly, when using TLC to sample loop conformations based on m consecutive tripeptides along a protein backbone, we obtain an exponential gain in the volume of the 5m‐dimensional configuration space to be explored. [ABSTRACT FROM AUTHOR]

The article deals with developing and testing a low-cost measuring system for simultaneous localisation and mapping (SLAM) in an indoor environment. The measuring system consists of three orthogonally-placed 2D lidars, a robotic platform with two wheel speed sensors, and an inertial measuring unit (IMU). The paper describes the data processing model used for both the estimation of the trajectory of SLAM and the creation of a 3D model of the environment based on the estimated trajectory of the SLAM. The main problem of SLAM usage is the accumulation of errors caused by the imperfect transformation of two scans into each other. The data processing developed includes an automatic evaluation and correction of the slope of the lidar. Furthermore, during the calculation of the trajectory, a repeatedly traversed area is identified (loop closure), which enables the optimisation of the trajectory determined. The system was tested in the indoor environment of the Faculty of Civil Engineering of the Slovak University of Technology in Bratislava.

The key constituent of simultaneous localization and mapping (SLAM) is the joint optimization of sensor trajectory estimation and 3D map construction. The multivariable optimization process in SLAM is mainly carried out through bundle adjustment (BA). However, the method of handling outliers in actual data directly affects the accuracy of BA optimization and further affects the tracking performance of the system. In addition, in monocular initialization process, when the feature points are coplanar or have low parallax, their fundamental matrix will degrade and greatly affects the initial pose estimation results. To further surmount the above challenges, this paper presents a real time monocular visual SLAM optimization based on BA with adaptive robust kernel (ARK-SLAM): 1) a model selection mechanism with geometric robust information content is designed to improve the robustness of monocular initialization; 2) an adaptive robust kernel based BA is proposed to reduce the interference of outliers and improve the accuracy of optimal pose estimation; 3) a loop closure candidate verification scheme based on adaptive robust kernel is introduced to jointly minimize the geometric error term and the relative pose constraints. The position tracking performance of developed ARK-SLAM method is verified using TUM RGB-D benchmark dataset and KITTI dataset, and experimental results illustrate the favorable performances of ARK-SLAM by comparing with ORB-SLAM, ORB-SLAM3, LSD-SLAM and PTAM. [ABSTRACT FROM AUTHOR]

This paper presents an approach to the estimation of a window shape for increasing the adaptability of glass façade-cleaning robots to different buildings. For this approach, a window scanning robot equipped with a 2D laser range scanner installed perpendicularly to a window surface is developed for the testbed, and a method for the window shape estimation is proposed, which consists of the robot's pose estimation with an extended Kalman filter (EKF) and the loop closure based on the robot's pose estimated. The effectiveness of the proposed approach is demonstrated through an experiment that is carried out on a window placed on a floor. The experimental results show that the window scanning robot can acquire a window shape, moving on a window surface, and the proposed approach is effective in increasing the accuracy of the window shape estimation. [ABSTRACT FROM AUTHOR]

Place recognition is an essential part of simultaneous localization and mapping (SLAM). LiDAR-based place recognition relies almost exclusively on geometric information. However, geometric information may become unreliable when faced with environments dominated by unstructured objects. In this paper, we explore the role of segmentation for extracting key structured information. We propose STV-SC, a novel segmentation and temporal verification enhanced place recognition method for unstructured environments. It contains a range image-based 3D point segmentation algorithm and a three-stage process to detect a loop. The three-stage method consists of a two-stage candidate loop search process and a one-stage segmentation and temporal verification (STV) process. Our STV process utilizes the time-continuous feature of SLAM to determine whether there is an occasional mismatch. We quantitatively demonstrate that the STV process can trigger false detections caused by unstructured objects and effectively extract structured objects to avoid outliers. Comparison with state-of-art algorithms on public datasets shows that STV-SC can run online and achieve improved performance in unstructured environments (Under the same precision, the recall rate is 1.4∼16% higher than Scan context). Therefore, our algorithm can effectively avoid the mismatching caused by the original algorithm in unstructured environment and improve the environmental adaptability of mobile agents. [ABSTRACT FROM AUTHOR]

This paper presents a new feature based approach for place recognition using 2D LiDAR (Light Detection And Ranging) data. The main contribution lies in the mapping process. It includes a keypoint selection strategy to model places with persistent keypoints from concatenated LiDAR scans. Our objective is to achieve map data reduction while maintaining good place recognition performace. LiDAR scans are concatenated with a registration algorithm and keypoints are extracted from each scan. Based on a regular grid, our approach measures the occurrence of similar keypoints in each region of interest defined by a grid cell. Only keypoints with occurrences beyond a threshold, qualified as stable keypoints, are kept in the place model called submap. The environment is therefore modeled as a collection of submaps, which constitutes the global map. Place recognition consists in submap matching followed by a two steps geometric verification. In the first stage, optimal parameters are set using corrected data. Mapping parameters satisfy six criteria among which is the spatial distribution distance, which represents another contribution of our work. It gives a measure on how well keypoints are distributed in space. Place recognition optimal parameters are set through global localization. In the second stage, the approach is evaluated using raw data in the contexts of global localization, and loop closure detection in a SLAM framework. The results obtained using popular real data sets show that our approach achieves significant map data reduction (up to $92\%$) while maintaining good place recognition performance, comparable to state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In this work, we investigate loop closure detection through a deep reinforcement learning approach. The loop closure detection problem correctly identifies a location or area a robot has previously visited. We propose a reward-driven optimization process that strives to learn loop closure detection. We demonstrate the framework in a simulated grid environment that generates observation data for a learning agent. We designed a grid-based environment to simulate indoor environments and train a policy for loop closure detection. A conversion of real-world map and features to the simulated environment is also demonstrated. The learning agent was tested in simulation and indoor lab environments. Our experimental results show that our algorithm can perform loop closure detection effectively. [ABSTRACT FROM AUTHOR]

Real-time 3D scene reconstruction has attracted a great amount of attention in the fields of augmented reality, virtual reality and robotics. Previous works usually assumed slow sensor motions to avoid large interframe differences and strong image blur, but this limits the applicability of the techniques in real cases. In this study, we propose an end-to-end 3D reconstruction system that combines color, depth and inertial measurements to achieve a robust reconstruction with fast sensor motions. We involved an extended Kalman filter (EKF) to fuse RGB-D-IMU data and jointly optimize feature correspondences, camera poses and scene geometry by using an iterative method. A novel geometry-aware patch deformation technique is proposed to adapt the changes in patch features in the image domain, leading to highly accurate feature tracking with fast sensor motions. In addition, we maintained the global consistency of the reconstructed model by achieving loop closure with submap-based depth image encoding and 3D map deformation. The experiments revealed that our patch deformation method improves the accuracy of feature tracking, that our improved loop detection method is more efficient than the original method and that our system possesses superior 3D reconstruction results compared with the state-of-the-art solutions in handling fast camera motions. [ABSTRACT FROM AUTHOR]

Simultaneous localization and mapping (SLAM) plays an important role in the area of robotics and augmented reality to simultaneously obtain its location and construct environment maps in real-time. There are many challenges in SLAM, such as data association, loop closure, and dynamic environments. In this paper, we propose a table retrieval method for SLAM data association and loop closure using semantic information in a dynamic environment. The detected landmarks are stored in a table for retrieval, and each landmark has its own semantic and location information for data association and loop closure. The proposed method only checks the corresponding items, so it is not necessary to traverse all the landmarks in the reference frames, which is beneficial to real-time performance and cost efficiency. Experiments are performed to verify the effectiveness of our method on the public TUM and KITTI dataset. The results show that our system achieves considerable performance improvement compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

In a number of technological operations, the end-effector of an open kinematic chain must follow a spatial trajectory, and a necessary condition for the realization of the movement is the smooth change of the generalized coordinates. An analogous problem exists in the development of drugs when a molecular configuration requiring minimal energy for its realization must be provided. A task of this nature must be solved in order to find the missing segments of a partially known protein structure. These problems lead to the more general task of ensuring a smooth change of the joint coordinates of the kinematic chain while performing a continuous spatial movement of the end-effector of the chain. An algorithm for planning a smooth reconfiguration of a redundant kinematics chain is proposed in the present work. The chain is decomposed into a countable number of substructures, each of which provides a unique solution of inverse kinematics for a target point. This makes it possible to select substructures during the spatial movement of the kinematic chain, ensuring a smooth change of the joint coordinates. The redundant generalized coordinates are force controlled, given the necessary values to ensure a smooth reconfiguration. The inverse kinematics problem is solved for the other joints. [ABSTRACT FROM AUTHOR]

In this article, we present a multiagent framework for real-time large-scale 3-D reconstruction applications. In SLAM, researchers usually build and update a 3-D map after applying nonlinear pose graph optimization techniques. Moreover, many multiagent systems are prevalently using odometry information from additional sensors. These methods generally involve extensive computer vision algorithms and are tightly coupled with various sensors. We develop a generic method for the key challenging scenarios in multiagent 3-D mapping based on different camera systems. The proposed framework performs actively in terms of localizing each agent after the first loop closure between them. It is shown that the proposed system only uses monocular cameras to yield real-time multiagent large-scale localization and 3-D global mapping. Based on the initial matching, our system can calculate the optimal scale difference between multiple 3-D maps and then estimate an accurate relative pose transformation for large-scale global mapping. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs) can experience significant performance issues during flight due to heavy CPU load, affecting their flight capabilities, communication, and endurance. To address this issue, this paper presents a lightweight stereo-inertial state estimator for addressing the heavy CPU load issue of ORB-SLAM. It utilizes nonlinear optimization and features to incorporate inertial information throughout the Simultaneous Localization and Mapping (SLAM) pipeline. The first key innovation is a coarse-to-fine optimization method that targets the enhancement of tracking speed by efficiently addressing bias and noise in the IMU parameters. A novel visual–inertial pose graph is proposed as an observer to assess error thresholds and guide the system towards visual-only or visual–inertial maximum a posteriori (MAP) estimation accordingly. Furthermore, this paper introduces the incorporation of inertial data in the loop closure thread. The IMU data provide displacement direction relative to world coordinates, which is serving as a necessary condition for loop detection. The experimental results demonstrate that our method maintains excellent localization accuracy compared to other state-of-the-art approaches on benchmark datasets, while also significantly reducing CPU load.

Simultaneous localization and mapping (SLAM) is a fundamental requirement for mobile robots like self-driving cars. Vision-based methods have advantages in sensor cost and loop closure detection, but are sensitive to illumination change and texture deficiency. Lidar-based SLAM systems perform better in accuracy, field-of-view and robustness to environmental changes, but may easily fail in structure-less scenarios. To compensate for the deficiencies of standalone sensors and provide more efficient SLAM functions, in this paper we propose a tightly coupled monocular visual lidar odometry, which tightly fuses the measurements of a monocular camera and a 3D lidar in a joint optimization. The system starts with a data preprocessing module, which outputs 3D visual and laser features through feature extraction and data association. Furthermore, the tightly coupled visual lidar odometry tightly fuses the visual and laser features in a unified optimization framework to estimate the transformation between consecutive scans. Finally, we combine visual and vicinity loop detection to construct loop constraints and optimize a 6-DOF global pose graph to achieve global-consistent pose estimation and environment mapping. The performance of our system is verified in the public KITTI dataset, and the experimental results demonstrate that the proposed method can run in real time with the 64-line lidar data and achieve better in accuracy, runtime and mapping against other state-of-the-art lidar-based and fusion-based methods. [ABSTRACT FROM AUTHOR]

This paper presents an approach to the estimation of a window shape for increasing the adaptability of glass façade-cleaning robots to different buildings. For this approach, a window scanning robot equipped with a 2D laser range scanner installed perpendicularly to a window surface is developed for the testbed, and a method for the window shape estimation is proposed, which consists of the robot’s pose estimation with an extended Kalman filter (EKF) and the loop closure based on the robot’s pose estimated. The effectiveness of the proposed approach is demonstrated through an experiment that is carried out on a window placed on a floor. The experimental results show that the window scanning robot can acquire a window shape, moving on a window surface, and the proposed approach is effective in increasing the accuracy of the window shape estimation.

Simultaneous localization and mapping (SLAM) is the essential technique in mapping environments that are denied to the global navigation satellite systems (GNSSs), such as indoor spaces. In this article, we present a loop-closing continuous-time LIDAR-IMU SLAM for indoor environments. The design of the proposed SLAM is based on arbitrarily-oriented planar features that allow for point to plane matching for local but also global optimization. Moreover, to update the SLAM graph during the optimization, we propose a simple yet elegant loop closure method in the form of merging the planes together. Representing the SLAM map by planes is advantageous due to the abundant existence of planar structures in indoor built environments. The proposed method was validated on a specific configuration of three 2D LIDAR scanners mounted on a wearable platform (backpack). Scanned point clouds were compared against ones obtained from a commercial mobile mapping system (Viametris iMS3D) and a terrestrial laser scanner (RIEGL VZ-400). The experimental results demonstrate that our SLAM system is capable of mapping multi-storey buildings, staircases, cluttered areas and areas with curved walls. Furthermore, our SLAM system offers comparable performance to that of the commercial system as shown by the low deviation between the point clouds generated by both systems. The majority of the cloud-to-cloud absolute distances – about 92% – are less than 3 cm. [ABSTRACT FROM AUTHOR]

This paper studies the folding of square, kirigami-inspired space structures consisting of concentrically arranged modular elements formed by thin shells. Localized elastic folds are introduced in the thin shells and different folding strategies can be obtained by varying the location of the folds and the sequence of imposed rotations. Modeling each modular element with rigid rods connected by revolute joints, numerical simulations of the kinematics of folding are obtained, including constraints that represent folding aids and a gravity offload system. These simulations are used to study two specific packaging schemes, and the folding envelopes of a specific structure are analyzed to identify the scheme that is easier to implement in practice. This particular scheme is demonstrated by means of a physical prototype. • We study the folding of space structures consisting of concentric modular elements. • Different packaging schemes are obtained by varying the locations of localized elastic folds. • A numerical model, based on rigid rods with revolute joints, is used to simulate the folding kinematics. • Specific packaging schemes that can be achieved with simple folding aids and gravity offloads are studied. • The scheme that is easier to implement is selected for an experimental demonstration. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles offer a safe and fast approach to the production of three-dimensional spatial data on the surrounding space. In this article, we present a low-cost SLAM-based drone for creating exploration maps of building interiors. The focus is on emergency response mapping in inaccessible or potentially dangerous places. For this purpose, we used a quadcopter microdrone equipped with six laser rangefinders (1D scanners) and an optical sensor for mapping and positioning. The employed SLAM is designed to map indoor spaces with planar structures through graph optimization. It performs loop-closure detection and correction to recognize previously visited places, and to correct the accumulated drift over time. The proposed methodology was validated for several indoor environments. We investigated the performance of our drone against a multilayer LiDAR-carrying macrodrone, a vision-aided navigation helmet, and ground truth obtained with a terrestrial laser scanner. The experimental results indicate that our SLAM system is capable of creating quality exploration maps of small indoor spaces, and handling the loop-closure problem. The accumulated drift without loop closure was on average 1.1% (0.35 m) over a 31-m-long acquisition trajectory. Moreover, the comparison results demonstrated that our flying microdrone provided a comparable performance to the multilayer LiDAR-based macrodrone, given the low deviation between the point clouds built by both drones. Approximately 85 % of the cloud-to-cloud distances were less than 10 cm.

Loop closure detection plays a vital role in the visual simultaneous localization and mapping (SLAM) systems. In order to overcome the shortcomings of the artificial design algorithm to extract insufficient features, this paper proposes a graph-regularization stacked denoising auto-encoder (G-SDAE) network that achieves high detection accuracy and improve reliability. This method is based on the SDAE and the manifold learning graph regularization structure. The G-SDAE preserves the local abstract geometry structure between features through spatial mapping in manifold learning, and the G-SDAE network can automatically extract abstract features, avoiding relying on empirical design algorithms to extract low-quality visual features. Compared with the bag-of-words (BoW) method, the OpenFABMAP algorithm, and the traditional SDAE method, extensive experiments show that the proposed algorithm achieves superior performances and provides a feasible solution for the loop closure detection part of the visual SLAM.

In this paper, we propose a stereo simultaneous localization and mapping (SLAM) method based on line segments. For the front-end module of SLAM, we designed a novel method based on the coplanar junction detection, description, and matching. Then the junctions along with their multi-scale rotated BRIEF descriptors are used in other SLAM modules, including line tracking, mapping, and loop closure. The line extraction and matching thread runs at 20 ~ 40Hz for stereo image sequences on a laptop, making it a practical front-end for line-based SLAM system. For the back-end module, a cost function is designed to minimize both of the reprojection error of line segments and alignment error of the vanishing points. The experimental results demonstrate that the proposed method exhibits more accurate localization and reconstruction than state-of-the-art line-based SLAM systems in line-rich environments.

Visual scene understanding and place recognition are the most challenging problems that mobile robots must solve for to achieve autonomous navigation. To reduce the high computational complexity of many global optimal search strategies, a new two-stage loop closure detection (LCD) strategy is developed in this paper. The front-end sequence node level matching (FSNLM) algorithm is based on the local continuity constraint of the motion process, which avoids the blind search for the global optimal match, and matches the image nodes via a sliding window to accurately find the local optimal matching candidate nodesets. In addition, the back-end image level matching (BILM) algorithm combined with an improved semantic model, DeepLab_AE, uses a convolutional neural network (CNN) as a feature detector to extract visual descriptors. It replaces traditional feature detectors that are manually designed by researchers in the computer vision field and cannot be applied to all environments. Finally, the performance of the two-stage LCD algorithm is evaluated on five public datasets, and is compared with the performance of other state-of-the-art algorithms. The evaluation results prove that the proposed method compares favorably with other algorithms. [ABSTRACT FROM AUTHOR]

Three-dimensional (3D) maps are indispensable for autonomous robot navigation in outdoor environments. Loop closure, which is a key technology in robotic mapping, is essential for creating consistent maps. This paper proposes an efficient method of loop detection for 3D-mapping. The proposed method detects loop constraints by estimating robot poses in revisited places using point-cloud registration. A difficulty in the registration is a large set of 3D points obtained by laser sensors, which requires a long processing time. To reduce the processing time, the proposed method employs a coarse-to-fine approach. Coarse estimation is performed using planes, lines, and balls instead of 3D points, and reduces the hypothesized loop constraints using geometric constraints between the segments. Subsequently, fine estimation is performed using the iterative closest points (ICP) algorithm and 3D points. Another difficulty is the precision of loop detection. To increase the precision, the proposed method employs robustification techniques such as outlier removal in the registration, combination of feature-based and pose-based methods, and robust pose adjustment. Experiments using large-scale datasets show that the proposed method realizes realtime loop detection in a variety of outdoor environments including cities, parks, and forest areas. [ABSTRACT FROM AUTHOR]

Loop closure detection is essential for simultaneous localization and mapping systems to decrease accumulating drift of trajectory estimations. Robust loop closure detection is specifically important in large-scale mapping, but it gets more challenging as the mapping environment grows. This paper proposes a SLAM system utilizing a two-pass loop closure detection method to improve mapping accuracy in large-scale environments. The proposed system finds loop closure candidates by employing global and local image features together. After selecting a group of candidates by similarity of global features, the system applies keypoint matching on this group to improve scene matching accuracy and determines loop closure candidates. We extensively evaluate the system on the widely used TUM RGB-D dataset, which contains sequences of small to large-scale indoor environments, with respect to different parameter combinations. The results show that the proposed method increases accuracy substantially and achieves large-scale mapping with acceptable overhead. [ABSTRACT FROM AUTHOR]

The problem of tripeptide loop closure is formulated in terms of the angles {Ti}(i)(3)(=1), describing the orientation of each peptide unit about the virtual axis joining the C alpha atoms. Imposing the constraint that at the junction of two such units the bond angle between the bonds C alpha-N and C alpha-C is fixed at some prescribed value 0 results in a system of three bivariate polynomials in u(i) : = tan tau(i)/2 of degree 2 in each variable. The system is analyzed for the existence of common solutions by making use of resultants, determinants of matrices composed of the coefficients of two (or more) polynomials, whose vanishing is a necessary and sufficient condition for the polynomials to have a common root. Two resultants are compared: the classical Sylvester resultant and the Dixon resultant. It is shown that when two of the variables are eliminated in favor of the third, a polynomial of degree 16 results. To each one of its real roots, there is a corresponding common zero of the system. To each such zero there corresponds a consistent conformation of the chain. The Sylvester method can find these zeros among the eigenvalues of a 24 X 24 matrix. For the Dixon approach, after removing extraneous factors, an optimally sized eigenvalue problem of size 16 X 16 results. Finally, the easy extension to the more general problem of triaxial loop closure is presented and an algorithm for implementing the method on arbitrary chains is given. (c) 2005 Wiley Periodicals, Inc.