A novel relocation method for simultaneous localization and mapping based on deep learning algorithm

Cao, Jun Hai; Zeng, Bi; Liu, Jianqi; Zhao, Zhenting; Su, Yongfeng

doi:10.1016/j.compeleceng.2017.03.015

Cited by 9 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, with the introduction of more complex and better models, how to ensure the real-time performance of model calculation? How to better set in the loop closure detection model in resource-constrained platforms, and the lightweight of the model is also a major problem [171].…”

Section: Key-frame Initialization Pose Graph Optimization Frame-wise ...mentioning

confidence: 99%

An Overview on Visual SLAM: From Tradition to Semantic

et al. 2022

View full text Add to dashboard Cite

Visual SLAM (VSLAM) has been developing rapidly due to its advantages of low-cost sensors, the easy fusion of other sensors, and richer environmental information. Traditional visionbased SLAM research has made many achievements, but it may fail to achieve wished results in challenging environments. Deep learning has promoted the development of computer vision, and the combination of deep learning and SLAM has attracted more and more attention. Semantic information, as high-level environmental information, can enable robots to better understand the surrounding environment. This paper introduces the development of VSLAM technology from two aspects: traditional VSLAM and semantic VSLAM combined with deep learning. For traditional VSLAM, we summarize the advantages and disadvantages of indirect and direct methods in detail and give some classical VSLAM open-source algorithms. In addition, we focus on the development of semantic VSLAM based on deep learning. Starting with typical neural networks CNN and RNN, we summarize the improvement of neural networks for the VSLAM system in detail. Later, we focus on the help of target detection and semantic segmentation for VSLAM semantic information introduction. We believe that the development of the future intelligent era cannot be without the help of semantic technology. Introducing deep learning into the VSLAM system to provide semantic information can help robots better perceive the surrounding environment and provide people with higher-level help.

show abstract

Section: Key-frame Initialization Pose Graph Optimization Frame-wise ...mentioning

confidence: 99%

An Overview on Visual SLAM: From Tradition to Semantic

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Kalman filtering or Monte Carlo-based particle filtering is usually used to obtain localization coordinates [12][13][14]. Laser radar positioning targets need to be achieved by identifying target features [15], for example, Ehsan J selected characteristics were buildings in cities [16]; Roelens J extracted geometric features to distinguish ditches [17]; Wang Jia identified tree species through six tree-shaped features [18]; Du S combined features based on points and grids to identify buildings [19]; Michael J successfully screened rough wood debris using height thresholds [20]; Yadav M used the Hough transform to identify the power line [21].…”

Section: Introductionmentioning

confidence: 99%

A Tanker Port Positioning Method of Quantitative Loading Automation

Zhu¹,

Ni²,

Huang³

et al. 2019

MCS

View full text Add to dashboard Cite

Laser scanning ranging radar is an important tool for machines to perceive the surrounding environment and is widely used in the power, forestry, surveying and mapping industries. At present, the loading of oil and grain oil in our country generally adopts the way of manual loading. The loading arm is inserted into the tank of the tanker for refueling, and the loading operation is very frequent. In order to realize automatic control of grain and oil loading, radar is needed to assist the robot to locate the oil port of the tanker. In this paper, a 360-degree laser scanning ranging radar is used to collect characteristic data of oil hole of tanker for the first time in simulated environment. Cubic spline interpolation was used to smooth and correct the radar scan data. Based on the feature that the distance data of oil port will change rapidly, an edge feature recognition algorithm is proposed to screen and calculate the target point, and then convert it to cartesian coordinate point, which can be used as the positioning target of the robot unit of quantitative loading system. The experimental results show that the method can locate the center of the circle accurately and meet the requirement of feature recognition accuracy.

show abstract

A Deep-Learning-based Strategy for Kidnapped Robot Problem in Similar Indoor Environment

Yan

Zhuang

et al. 2020

J Intell Robot Syst

View full text Add to dashboard Cite

A novel relocation method for simultaneous localization and mapping based on deep learning algorithm

Cited by 9 publications

References 16 publications

An Overview on Visual SLAM: From Tradition to Semantic

An Overview on Visual SLAM: From Tradition to Semantic

A Tanker Port Positioning Method of Quantitative Loading Automation

A Deep-Learning-based Strategy for Kidnapped Robot Problem in Similar Indoor Environment

Contact Info

Product

Resources

About