Compared to diverse feature detectors and descriptors used for visual scenes, describing a place using structural information is relatively less reported. Recent advances in simultaneous localization mapping (SLAM) provides dense 3D maps of the environment proposed by sensors. Toward global based on information, we propose Scan Context, non-histogram-based descriptor from Light Detection Ranging (LiDAR) scans. Unlike previously reported methods, approach directly records structure visible...

Inspection of ship hulls and marine structures using autonomous underwater vehicles has emerged as a unique challenging application robotics. The problem poses rich questions in physical design operation, perception navigation, planning, driven by difficulties arising from the acoustic environment, poor water quality highly complex to be inspected. In this paper, we develop apply algorithms for central navigation planning problems on hulls. These divide into two classes, suitable open,...

This paper reports a real-time monocular visual simultaneous localization and mapping (SLAM) algorithm results for its application in the area of autonomous underwater ship hull inspection. The proposed overcomes some specific challenges associated with SLAM, namely, limited field view imagery feature-poor regions. It does so by exploiting our SLAM navigation prior within image registration pipeline being selective about which is considered informative terms map. A novel online bag-of-words...

The high diversity of urban environments, at both the inter and intra levels, poses challenges for robotics research. Such include discrepancies in features between cities deterioration sensor measurements within a city. With such consideration, this paper aims to provide Light Detection Ranging (LiDAR) image data acquired complex environments. In contrast existing datasets, presented dataset encapsulates various addresses major issues areas, as unreliable sporadic Global Positioning System...

This paper introduces a multimodal range dataset namely for radio detection and ranging (radar) light (LiDAR) specifically targeting the urban environment. By extending our workshop [1] to larger scale, this focuses on sensor-based place recognition provides 6D baseline trajectories of vehicle ground truth. Provided radar data support both raw-level image-format data, including set time-stamped 1D intensity arrays 360° polar images, respectively. In doing so, we provide flexibility between...

Place recognition is a key module in robotic navigation. The existing line of studies mostly focuses on visual place to recognize previously visited places solely based their appearance. In this article, we address structural by recognizing appearance, namely from range sensors. Extending our previous work rotation invariant spatial descriptor, the proposed descriptor completes generic robust both (heading) and translation when roll–pitch motions are not severe. We introduce two...

Abstract Odometry is crucial for robot navigation, particularly in situations where global positioning methods like system are unavailable. The main goal of odometry to predict the robot’s motion and accurately determine its current location. Various sensors, such as wheel encoder, inertial measurement unit (IMU), camera, radar, Light Detection Ranging (LiDAR), used robotics. LiDAR, particular, has gained attention ability provide rich three-dimensional (3D) data immunity light variations....

This paper reports on an integrated navigation algorithm for the visual simultaneous localization and mapping (SLAM) robotic area coverage problem. In problem, goal is to explore map a given target within reasonable amount of time. necessitates use minimally redundant overlap trajectories efficiency; however, SLAM’s estimate will inevitably drift over time in absence loop closures. Therefore, efficient good SLAM performance represent competing objectives. To solve this decision-making we...

In this letter, we present a long-term localization method that effectively exploits the structural information of an environment via image format. The proposed presents robust year-round performance even when learned in just single day. localizer learns point cloud descriptor, named Scan Context Image (SCI), and performs robot on grid map by formulating place recognition problem as classification using convolutional neural network. Our is faster than existing methods for because it avoids...

We present a novel static point cloud map construction algorithm, called Removert, for use within dynamic urban environments. Leaving only points and excluding objects is critical problem in various robust robot missions changing outdoors, the procedure commonly contains comparing query to noisy that has points. In doing so, however, estimated discrepancies between scan tend possess errors due imperfect pose estimation, which degrades quality. To tackle problem, we propose multiresolution...

This paper describes a framework for direct visual simultaneous localization and mapping (SLAM) combining monocular camera with sparse depth information from Light Detection Ranging (LiDAR). To ensure realtime performance while maintaining high accuracy in motion estimation, we present (i) sliding window-based tracking method, (ii) strict pose marginalization accurate pose-graph SLAM (iii) depth-integrated frame matching large-scale mapping. Unlike conventional feature-based LiDAR mapping,...

This paper presents a fast single image enhancement that is applicable regardless of channels in various environments. The main idea the combining model-based and fusion-based dehazing methods, thereby presenting balanced while elaborating details. proposed method enhances both color grayscale images without any prior information. Multiband decomposition utilized to extract base detail layers for intensity Laplacian modules. ambient map transmission estimation module are effective restoring...

Learning-based ego-motion estimation approaches have recently drawn strong interest from researchers, mostly focusing on visual perception. A few learning-based using Light Detection and Ranging (LiDAR) been re-ported; however, they heavily rely a supervised learning manner. Despite the meaningful performance of these approaches, training requires ground-truth pose labels, which is bottleneck for real-world applications. Differing we focus unsupervised LiDAR odometry (LO) without trainable...

Recent studies in radar-based navigation present promising performance using scanning radars. These odometry methods are mostly feature-based; they detect and match salient features within a radar image. Differing from existing feature-based methods, this paper reports on method direct odometry, PhaRaO, which infers relative motion pair of scans via phase correlation. Specifically, we apply the Fourier Mellin transform (FMT) for Cartesian log-polar images to sequentially estimate rotation...

In this letter, we present a dataset capturing diverse visual data formats that target varying luminance conditions. While RGB cameras provide nourishing and intuitive information, changes in lighting conditions potentially result catastrophic failure for robotic applications based on vision sensors. Approaches overcoming illumination problems have included developing more robust algorithms or other types of sensors, such as thermal event cameras. Despite the alternative sensors' potential,...

This paper reports on an integrated navigation algorithm for the visual simultaneous localization and mapping (SLAM) robotic area coverage problem. In problem, goal is to explore map a given target in reasonable amount of time. necessitates use minimally redundant overlap trajectories efficiency; however, SLAM's estimate will inevitably drift over time absence loop-closures. Therefore, efficient good SLAM performance represent competing objectives. To solve this decision-making we introduce...

2D nanoscale oxides have attracted a large amount of research interest due to their unique properties. Here, facile synthetic approach prepare graphene‐mimicking, porous Co 3 O 4 nanofoils using graphene oxide (GO) as sacrificial template is reported. The thermal instability graphene, well the catalytic ability particles degrade carbon backbones, allow fabrication without loss nature GO. Based on these results, mimicking route for large‐area transition metal applications in electrochemical...

This paper reports on a system for an autonomous underwater vehicle to perform in situ , multiple session hull inspection using long‐term simultaneous localization and mapping (SLAM). Our method assumes very little priori knowledge, it does not require the aid of acoustic beacons navigation, which is typical mode navigation this type application. combines recent techniques saliency‐informed visual SLAM representing ship surface as collection many locally planar features. methodology produces...

This paper presents a method for effective ambient light and transmission estimation in underwater images using common convolutional network architecture. The estimated the map are used to dehaze images. Dehazing is especially challenging due unknown significantly varying environments. Unlike dehazing methods, proposed capable of estimating along with thereby improving reconstruction quality dehazed We evaluate performance on real also compare our current state-of-the-art techniques.

In this paper, we propose the Road-SLAM algorithm, which robustly exploits road markings obtained from camera images. Road are well categorized and informative but susceptible to visual aliasing for global localization. To enable loop-closures using marking matching, our method defines a feature consisting of surrounding lanes as sub-map. The proposed uses random forest improve accuracy matching sub-map containing information. classifies into six classes only incorporates avoid ambiguity. is...

In this letter, we propose a thermal-infrared simultaneous localization and mapping (SLAM) system enhanced by sparse depth measurements from light detection ranging (LiDAR). Thermal-infrared cameras are relatively robust against fog, smoke, dynamic lighting conditions compared to RGB operating under the visible spectrum. Exploiting for motion estimation is highly appealing. However, utilizing camera directly in existing vision-based methods difficult because of modality difference. This...

Place recognition is crucial for robot localization and loop closure in simultaneous mapping (SLAM). Light Detection Ranging (LiDAR), known its robust sensing capabilities measurement consistency even varying illumination conditions, has become pivotal various fields, surpassing traditional imaging sensors certain applications. Among types of LiDAR, spinning LiDARs are widely used, while non-repetitive scanning patterns have recently been utilized robotics Some provide additional...

Aqueous rechargeable Zn batteries (AZBs) are considered to be promising next-generation battery systems. However, the growth of dendrites and water-induced side reactions have hindered their practical application, especially with regard long-term cyclability. To address these challenges, we introduce a supramolecular metal-organic framework (SMOF) coating layer using an α-cyclodextrin-based MOF (α-CD-MOF-K) polymeric binder. The plate-like α-CD-MOF-K particles, combined binder create dense...