Real-Time Lidar Odometry and Mapping with Loop Closure

Liu, Yonghui; Zhang, Weimin; Li, Fangxing; Zuo, Zhengqing; Huang, Qiang

doi:10.3390/s22124373

Cited by 4 publications

(2 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For each data acquisition scene, based on the first data collected, the LiDAR Odometry and Mapping (LOAM) algorithm [ 46 ] is used to construct an a priori point cloud map. Because the secondary collected LiDAR point cloud encompasses many dynamic targets, errors will arise when directly detecting dynamic targets directly using the scan-to-map approach [ 47 ] to register a single-frame point cloud and a point cloud map. To obtain a more accurate training set classification label, the descriptor comparison method is used to detect the motion state of each point cloud cluster in each frame of the LiDAR point cloud by dividing the grid and comparing the descriptors based on the registered single-frame point cloud and the prior map.…”

Section: Methodsmentioning

confidence: 99%

LiDAR Dynamic Target Detection Based on Multidimensional Features

Xu,

Gao,

Sui

et al. 2024

Sensors

View full text Add to dashboard Cite

To address the limitations of LiDAR dynamic target detection methods, which often require heuristic thresholding, indirect computational assistance, supplementary sensor data, or postdetection, we propose an innovative method based on multidimensional features. Using the differences between the positions and geometric structures of point cloud clusters scanned by the same target in adjacent frame point clouds, the motion states of the point cloud clusters are comprehensively evaluated. To enable the automatic precision pairing of point cloud clusters from adjacent frames of the same target, a double registration algorithm is proposed for point cloud cluster centroids. The iterative closest point (ICP) algorithm is employed for approximate interframe pose estimation during coarse registration. The random sample consensus (RANSAC) and four-parameter transformation algorithms are employed to obtain precise interframe pose relations during fine registration. These processes standardize the coordinate systems of adjacent point clouds and facilitate the association of point cloud clusters from the same target. Based on the paired point cloud cluster, a classification feature system is used to construct the XGBoost decision tree. To enhance the XGBoost training efficiency, a Spearman’s rank correlation coefficient-bidirectional search for a dimensionality reduction algorithm is proposed to expedite the optimal classification feature subset construction. After preliminary outcomes are generated by XGBoost, a double Boyer–Moore voting-sliding window algorithm is proposed to refine the final LiDAR dynamic target detection accuracy. To validate the efficacy and efficiency of our method in LiDAR dynamic target detection, an experimental platform is established. Real-world data are collected and pertinent experiments are designed. The experimental results illustrate the soundness of our method. The LiDAR dynamic target correct detection rate is 92.41%, the static target error detection rate is 1.43%, and the detection efficiency is 0.0299 s. Our method exhibits notable advantages over open-source comparative methods, achieving highly efficient and precise LiDAR dynamic target detection.

show abstract

Section: Methodsmentioning

confidence: 99%

LiDAR Dynamic Target Detection Based on Multidimensional Features

Xu,

Gao,

Sui

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…The feature point explicit matching technique is used in most image-based localization methods. Such approaches to localization typically use the structure from motion (SFM) [22] or SLAM techniques [23,24] to represent the scene by obtaining a 3D model. This algorithm specifies the robot position of the image be queried from a series of 2D-3D correspondences.…”

Section: Introductionmentioning

confidence: 99%

An End‐to‐End Robotic Visual Localization Algorithm Based on Deep Learning

Chen,

Wang,

Fan

et al. 2023

Journal of Sensors

View full text Add to dashboard Cite

Efficient localization plays a significant role in mobile autonomous robots’ navigation systems. Traditional visual simultaneous localization systems based on point feature matching suffer from two shortcomings. First one is that the method of tracking features is not robust for environments with frequent changes in brightness. Another one is the large of consecutive visual keyframes can consume expensive computational and storage resources in complex environments. To solve these problems, an end-to-end real-time six degrees of freedom object pose estimation algorithm is proposed to solve the robust and efficient challenges through a deep learning model. First, preprocessing operations such as cropping, averaging, and timestamp alignment are performed on datasets to reduce computational cost and time. Second, the processed dataset is fed into our neural network model to extract the most effective features for matching. Finally, the robot’s current 3D translation and 4D angle information are predicted and output to achieve an end-to-end localization system. A broad range of experiments are performed on both indoor and outdoor datasets. The experimental results demonstrate that the translation and orientation accuracy in outdoor scenes improved by 32.9% and 31.4%, respectively. The average improvement of localization accuracy in indoor scenes is 38.4%, and the angle improvement is 13.1%. Moreover, the effectiveness of predicting the global motion trajectories of sequential images algorithm has been verified and is superior to other convolutional neural network methods.

show abstract