A local Bundle Adjustment (BA) on a sliding window of keyframes has been widely used in visual SLAM and proved to be very effective lowering the drift. But lidar SLAM, BA method is hardly because sparse feature points (e.g., edge plane) make exact point matching impossible. In this letter, we formulate as minimizing distance from its matched or plane. Unlike (and prior plane adjustment SLAM) where co-determined along with pose, show that can analytically solved removed BA, resultant only...

High performance light detection and ranging (lidars) are essential in autonomous robots such as self-driving cars, automated ground vehicles intelligent machines. Traditional mechanical scanning lidars offer superior vehicles, but the potential mass application is limited by inherent manufacturing difficulty. In this article, we propose a robotic lidar sensor based on incommensurable that allows straightforward production adoption robots. Some unique features additionally permitted...

Simultaneous determination of sensor poses and scene geometry is a fundamental problem for robot vision that often achieved by Bundle Adjustment (BA). This article presents an efficient consistent bundle adjustment method light detection ranging (lidar) sensors. The employs edge plane features to represent the geometry, directly minimizes natural Euclidean distance from each raw point respective feature. A nice property this formulation can be analytically solved, drastically reducing...

Accurate and robust place recognition is essential for robot navigation, yet achieving full pose invariance high performance across diverse scenes remains challenging. In this work, we propose a novel global local combined descriptor named Binary Triangle Combined (BTC) descriptor. We first extract the keypoints of point cloud by projecting points to planes extracted therein. Any three form unique triangle, with lengths its sides constituting triangle that captures appearance cloud. Thanks...

Reconstructing an accurate and consistent large-scale LiDAR point cloud map is crucial for robotics applications. The existing solution, pose graph optimization, though it time-efficient, does not directly optimize the mapping consistency. bundle adjustment (BA) has been recently proposed to resolve this issue; however, too time-consuming on maps. To mitigate problem, paper presents a globally efficient method suitable Our work consists of bottom-up hierarchical BA top-down which combines...

Owing to the possible existence of system failures and packet dropouts in wastewater treatment process (WWTP), it is difficult obtain sufficient data, which will result data shortage. Therefore, a challenge design an effective data-driven controller with above shortage issue for WWTP. To solve this problem, syncretic fuzzy-neural (SFNC) was developed analyzed article. First, knowledge obtained from operation conditions made full use by reconstruction mechanism construct initial condition...

Wind power is one of the majority sustainable sources in world. However, wind turbine facing highly possible blade frozen event on blade. While numerous ice detection methods have been reported, with data acquired by SCADA (supervisory control and acquisition) system has not investigated yet. Thus, this study we propose a assessment framework using an ITL (inductive transfer learning) approach. We applied different machine learning to problem, which fully-connected neural networks (FNN)...

It is well known that extracorporeal shock wave treatment capable of providing a non-surgical and relatively pain free alternative modality for patients suffering from musculoskeletal disorders but do not respond to conservative treatments. The major objective current work investigate how the (SW) field would change if bony structure exists in path acoustic wave. Here, model finite element method (FEM) was developed based on linear elasticity propagation equations examine SW deflection near...

Abstract A turbocharger turbine is exposed to pulsating flow conditions when it connected an engine exhaust system due the opening and closing of valves. However, many radial turbines are designed tested under steady-state without taking into account these unsteady flows. In order seek optimal aerodynamic design a (RFT) conditions, present research utilizes numerical simulation approach optimize blade shape small-scale mixed (MFT) 50 Hz pulses. This corresponds four-stroke, three-cylinder...

This paper introduces a novel targetless method for joint intrinsic and extrinsic calibration of LiDAR-camera systems using plane-constrained bundle adjustment (BA). Our leverages LiDAR point cloud measurements from planes in the scene, alongside visual points derived those planes. The core novelty our lies integration BA with registration between planes, which is formulated as unified optimization problem. formulation achieves concurrent calibration, while also imparting depth constraints...

Traditional renal puncture surgery requires manual operation, which has a poor effect, low surgical success rate, and high incidence of postoperative complications. Robot-assisted can effectively improve the accuracy punctures, rate surgery, reduce occurrence This paper provides dual-armed robotic scheme to assist surgeons. The system is divided into an ultrasound scanning arm arm. Both arms with compliant positioning function master–slave control are designed, respectively, achieved. arm’s...

Randomly encoded compressive sensing (CS) has potential in fast acquisition of magnetic resonance imaging (MRI) data most naturally compressible images. However, there is no guaranteed good performance for general applications by any the traditional CS-MRI theoretical schemes developed so far. On other hand, recent research demonstrates that adaptive sampling exploiting tree structure nonzero wavelet coefficients signals allows more control over procedure form feedback and improve CS...

Human-computer interface (HCI) is very important to improve the efficiency of robot control support complex tasks. It valuable develop recognition techniques by which compute can be capable handling humanpsilas continuous gesture effectively. The present work seeks model human motions in a manner for leaning and recognition. Toward this end,a novel HCI approach based on hidden Markov model(HMM) proposed semantically meaningful movements sequence camera images. As maximum likelihood...

Point cloud maps with accurate color are crucial in robotics and mapping applications. Existing approaches for producing RGB-colorized primarily based on real-time localization using filter-based estimation or sliding window optimization, which may lack accuracy global consistency. In this work, we introduce a novel LiDAR-Visual bundle adjustment (BA) named LVBA to improve the quality of RGB point beyond existing baselines. first optimizes LiDAR poses via BA, followed by photometric visual...

For the past few years, electroencephalogram (EEG)-based brain-computer interfaces (BCIs) have gotten tremendous progress and attracted increasing attention. To broaden application of BCIs, researchers focused on increasement BCI instruction number in recent years. However, with a large instructions, calibration time will be too long to accepted practical usage. This study proposed new coding method based multifocal steady-state visual evoked potentials (mfSSVEPs), which 16 targets were...

본 논문에서는 열림부분에 기반을 둔 자동차 번호판의 숫자인식 기법을 제안한다. 제안된 인식기법에서 숫자를 상부와 하부로 나누고, 각각에 대해 열림부분을 판정하여 인식한다. 기법에서, 하부의 분할은 정해진 절단선을 바탕으로 하며, 교점의 개수에 따라 절단선의 위치가 조절된다. 방법은 템플릿 매칭 방법에 비해 잡음이나 회전 등의 영향을 받지 않으며 강건하다. 모의실험에서 기법의 성능을 평가하기 위해 번호판에 사용되는 숫자들에 처리를 하고, 그 결과를 제시하였다. 기법은 숫자인식에서 매우 높은 인식률을 보인다. In this paper, we propose a new digit recognition method based on opened enclosure. the proposed method, each is divided into two parts, an upper part and lower part, which are cutting line that modified...

High performance lidars are essential in autonomous robots such as self-driving cars, automated ground vehicles and intelligent machines. Traditional mechanical scanning offer superior vehicles, but the potential mass application is limited by inherent manufacturing difficulty. We propose a robotic lidar sensor based on incommensurable that allows straightforward production adoption robots. Some unique features additionally permitted this scanning. Similar to fovea human retina, peaked...

Based on vision inspection technique, the measurement system for measuring complexity free form surface is developed. For getting high-accuracy object surface, subpixel subdivision technology studied. sake of data got from are scatted points, modeling technique triangular meshes should be used. The basic idea approach based Bezier interpolation can expressed. A practical algorithm reconstruction combining triangulation and NURBS rectangular proposed. In 3-D digital image merging depth images...