The registration of point clouds is a key ingredient LiDAR-based SLAM systems and mapping approaches. A challenging task in this context finding the right data association between 3D points. This paper proposes novel flexible graph network architecture to tackle keypoint matching problem an end-to-end fashion. Each layer our multiplex dynamic attention (MDGAT) utilizes mechanism dynamically construct reasons about contextual information based on cloud data. It enriches feature representation...

To isolate and characterize novel angiotensin I-converting enzyme (ACE) inhibitory peptide from loach (Misgurnus anguillicaudatus), six proteases, pepsin, α-chymotrypsin, bromelain, papain, alcalase, Neutrase, were used to hydrolyze protein. The hydrolysate (LPH) generated by bromelain [ratio of substrate, 3:1000 (w/w)] was found have the highest ACE activity (IC(50), 613.2 ± 8.3 μg/mL). Therefore, it treated ultrafiltration afford fraction LPH-IV (MW < 2.5 kDa) with an IC(50) 231.2 3.8...

This article presents a sliding widow-based visual-inertial odometry to deal with the micro air vehicle (MAV) pose estimation problem. Errors caused by inertial measurement unit (IMU) preintegration, visual landmarks reprojection, and marginalization, are unified into nonlinear residual minimization framework. Furthermore, dual-step marginalization method has been proposed increase computational efficiency. Experiments have conducted on publicly available datasets, as well customized...

In this paper, we propose a semantic simultaneous localization and mapping (SLAM) framework for rescue robots, report its use in navigation tasks. Our can generate not only geometric maps the form of dense point-clouds but also corresponding point-wise labels generated by segmentation convolutional neural network (CNN). The CNN is trained using our RGB-D dataset RoboCup Rescue-Robot-League (RRL) competition environment. With help information, robot identify different types terrains complex...

Most state estimation procedures in mobile robotics require information about the locations of individual sensors on platform. In this paper, we study motion-based multi-sensor extrinsic calibration problem and point out an overlooked defect traditional least squares context. We present a novel approach based-on Gauss-Helmert paradigm, together with formulation for motion constraint. Our estimates not only parameters but also pose observation errors, thus recovering underlying sensor...

Most autonomous mobile robots are often equipped with monocular cameras and 3D LiDARs to perform vital tasks such as localization mapping. In this paper, we present a two-stage extrinsic calibration method well hybrid-residual-based odometry approach for camera-LiDAR systems. Our can estimate the relative transformation between camera LiDAR high accuracy, allowing us better register image point cloud data. After calibration, our be used provide real-time, accurate estimates. exploits both...

In this paper, we address the problem of combining 3D laser scanner and camera information to estimate motion a mobile platform. We propose direct laser-visual odometry approach building upon photometric image alignment. Our is designed maximize usage both, scan, compute an accurate frame-to-frame estimate. To deal with sparsity range measurements, our identifies planar point sets within individual clouds subsequently extract their corresponding pixel patches from image. The extracted are...

Abstract Tracked robots operating on rough terrain are often equipped with controllable flippers to help themselves overcome large obstacles or gaps. How automate the control of these auxiliary achieve autonomous traversal remains an open question, which still necessitates inefficient manual teleoperation in practice. To tackle this problem, article presents a geometry‐based motion planning method for articulated tracked robot self‐control its during semiautonomous over urban search and...

Tracked robots equipped with flippers and LiDAR sensors have been widely used in urban search rescue. Achieving autonomous flipper control is important enhancing the intelligent operation of tracked within complex rescuing environments. While existing methods mainly rely on heavy work manual modeling, this paper proposes a novel Deep Reinforcement Learning (DRL) approach named ICM-D3QN for terrains. Specifically, comprises three modules: feature extraction fusion module extracting...

Purpose – The purpose of this paper is to design intelligent robots operating in such dynamic environments like the RoboCup Middle-Size League (MSL). In MSL, two teams five autonomous play on an 18- × 12-m field. Equipped with sensors and on-board computers, each robot should be able perceive environment, make decision control itself soccer game autonomously. Design/methodology/approach This presents our robots, participating MSL. mechanical platform, electrical architecture software...

Extrinsic sensor calibration is an important task in robotics. There are various ways to perform the task, but it often remains unclear which methods better than others. In this paper, we provide a systematic study about accuracy of three types methods, each represented by abstract geometric model based on configuration and setup. We discuss advantages disadvantages rigorous their noise sensitivity from perspective. As result, can reveal quantify relative accuracies models, thus answering...

Pose estimation and mapping are key capabilities of most autonomous vehicles thus a number localization SLAM algorithms have been developed in the past. Autonomous robots cars typically equipped with multiple sensors. Often, sensor suite includes camera laser range finder. In this paper, we consider problem incremental ego-motion estimation, using both, monocular finder jointly. We propose new algorithm, that exploits advantages both sensors-the ability cameras to determine orientations well...

Cordyceps militaris contains many kinds of bioactive components with the various physiological activities, but has less natural sources fruiting bodies. Compared bodies from same origin, fermented mycelia contained significantly higher amount protein, crude fat and polysaccharides. The had more total amino acids, proline (3.84 ± 0.12 g 100g<SUP>-1</SUP>), leucine (2.96 0.10 aspartic acid (2.68 0.06 100g<SUP>-1</SUP>) as top three aboundant while glutamic (3.56 0.07 (2.52 0.13 valine (2.35...

This paper describes the design, realization and testing of an active ball handling mechanism which is designed for dribbling in RoboCup Middle Size League used by our team NuBot. First, a virtual model simplified derived. Afterwards, control system including feedback controller feedforward developed step step. Experimental results show that stays contact with robot using proposed when driving forwards, backwards rotating.

With Light Detection and Ranging(LiDAR) technology's development maturity, various new LiDARs have emerged in recent years, such as the Mid-40 series launched by Livox. Compared with rotating Velodyne LiDAR, its small scanning range irregular trajectory pose challenges for SLAM applications. We evaluate characteristics of Mid-40, propose a calibration method based on checkerboard patterns. Our extracts corners' coordinates from both point cloud image, which are then used to establish 3D-2D...

Two-dimensional (2D) perovskites have been widely applied in photoelectric devices due to their excellent optical and electrical properties. However, the low absorption of ultrathin thickness inhibits photoelectron generation thus limits performances. Herein, we constructed 2D CsPbBr3 composite structures by combining them with g-C3N4 (CN) graphene oxide (GO). As compared devices, photocurrents GO/CsPbBr3 CN/CsPbBr3 increase 68 114%, respectively. Based on experimental spectra combination...

Abstract β-D-glucosidase (βG) gene from Oenococcus oeni SD-2a and 31MBR was cloned, sequenced analyzed, also intracellular βG of the two strains further localized. The results showed that in high homology (&gt; 99%) to reported gene, confirming both possess activity at molecular level. Intracellular is a mainly soluble protein, existing mostly cytoplasm some extent periplasm. While for 31MBR, insoluble protein cytoplasmic membrane. This study provides basic information metabolic mechanism O. 31MBR.

Crowd navigation has becoming an increasingly prominent problem in robotics. The main challenge comes from the lack of understanding pedestrians' behaviors. Encouraged by great achievement trajectory prediction, twin field crowd navigation, this work focus on integrating prediction with path planning and proposed a algorithm named RHC-T (Receding Horizon Control Trajectron++). It consists two independent modules: one for another receding horizon control. Benefiting module, builds up explicit...

In this letter, we supplement the derivation in Laurent's paper by mathematically proving that combining binary continuous phase modulation (CPM) signals different bit durations results decomposed CPM expression consisting of a set amplitude-modulated pulse functions.

The autonomous control of flippers plays an important role in enhancing the intelligent operation tracked robots within complex environments. While existing methods mainly rely on hand-crafted models, this paper, we introduce a novel approach that leverages deep reinforcement learning (DRL) techniques for flipper terrains. Specifically, propose new DRL network named AT-D3QN, which ensures safe and smooth robots. It comprises two modules, feature extraction fusion module extracting...