Polarized light-aided VINS: Global heading measurements and graph optimization based multi-sensor fusion

2021 ◽  
Author(s):  
Linlin Xia ◽  
Ruimin Liu ◽  
Daochang Zhang ◽  
Jingjing Zhang
Keyword(s):  
Sensor Fusion ◽  
Polarized Light ◽  
Visual Slam ◽  
Combined System ◽  
Polarization Model ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
Passive Navigation ◽  
Navigation Mode ◽  
Robot Platform

Abstract Polarized skylight is as fundamental a constituent of passive navigation as geomagnetic field. In regards to its applicability to outdoor robot localization, a polarized light-aided VINS (abbreviates ‘visual-inertial navigation system’) modelization dedicated to globally optimized pose estimation and heading correction is constructed. The combined system follows typical visual SLAM (abbreviates ‘simultaneous localization and mapping’) frameworks, and we propose a methodology to fuse global heading measurements with visual and inertial information in a graph optimization based estimator. With ideas of ‘new-added attribute of each vertex and heading error encoded constraint edges’, the heading, as absolute orientation reference, is estimated by Berry polarization model and continuously updated in a graph structure. The formulized graph optimization process for multi-sensor fusion is simultaneously provided. In terms of campus road experiments on Bulldog-CX Robot platform, results are compared against purely stereo camera-dependent and VINS Fusion frameworks, revealing our design is substantially more accurate than others with both locally and globally consistent position and attitude estimates. As essentially passive, anatomically coupled and drifts calibratable navigation mode, the polarized light-aided VINS may therefore be considered as a tool candidate for a class of visual SLAM based multi-sensor fusion.

scholarly journals A LiDAR/Visual SLAM Backend with Loop Closure Detection and Graph Optimization

2021 ◽  
Vol 13 (14) ◽  
pp. 2720
Author(s):  
Shoubin Chen ◽  
Baoding Zhou ◽  
Changhui Jiang ◽  
Weixing Xue ◽  
Qingquan Li
Keyword(s):  
Point Clouds ◽  
Visual Features ◽  
Visual Slam ◽  
Key Factors ◽  
Loop Closure ◽  
Loop Closure Detection ◽  
Active Sensor ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
Pose Graph Optimization

LiDAR (light detection and ranging), as an active sensor, is investigated in the simultaneous localization and mapping (SLAM) system. Typically, a LiDAR SLAM system consists of front-end odometry and back-end optimization modules. Loop closure detection and pose graph optimization are the key factors determining the performance of the LiDAR SLAM system. However, the LiDAR works at a single wavelength (905 nm), and few textures or visual features are extracted, which restricts the performance of point clouds matching based loop closure detection and graph optimization. With the aim of improving LiDAR SLAM performance, in this paper, we proposed a LiDAR and visual SLAM backend, which utilizes LiDAR geometry features and visual features to accomplish loop closure detection. Firstly, the bag of word (BoW) model, describing the visual similarities, was constructed to assist in the loop closure detection and, secondly, point clouds re-matching was conducted to verify the loop closure detection and accomplish graph optimization. Experiments with different datasets were carried out for assessing the proposed method, and the results demonstrated that the inclusion of the visual features effectively helped with the loop closure detection and improved LiDAR SLAM performance. In addition, the source code, which is open source, is available for download once you contact the corresponding author.

scholarly journals Map Construction Based on LiDAR Vision Inertial Multi-Sensor Fusion

2021 ◽  
Vol 12 (4) ◽  
pp. 261
Author(s):  
Chuanwei Zhang ◽  
Lei Lei ◽  
Xiaowen Ma ◽  
Rui Zhou ◽  
Zhenghe Shi ◽  
...  
Keyword(s):  
Sensor Fusion ◽  
Point Cloud ◽  
Factor Graph ◽  
Mapping Accuracy ◽  
Map Construction ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
Measurement Units ◽  
Screening Condition ◽  
Abnormal Points

In order to make up for the shortcomings of independent sensors and provide more reliable estimation, a multi-sensor fusion framework for simultaneous localization and mapping is proposed in this paper. Firstly, the light detection and ranging (LiDAR) point cloud is screened in the front-end processing to eliminate abnormal points and improve the positioning and mapping accuracy. Secondly, for the problem of false detection when the LiDAR is surrounded by repeated structures, the intensity value of the laser point cloud is used as the screening condition to screen out robust visual features with high distance confidence, for the purpose of softening. Then, the initial factor, registration factor, inertial measurement units (IMU) factor and loop factor are inserted into the factor graph. A factor graph optimization algorithm based on a Bayesian tree is used for incremental optimization estimation to realize the data fusion. The algorithm was tested in campus and real road environments. The experimental results show that the proposed algorithm can realize state estimation and map construction with high accuracy and strong robustness.

scholarly journals Research on Visual Positioning of a Roadheader and Construction of an Environment Map

2021 ◽  
Vol 11 (11) ◽  
pp. 4968
Author(s):  
Wentao Zhang ◽  
Guodong Zhai ◽  
Zhongwen Yue ◽  
Tao Pan ◽  
Ran Cheng
Keyword(s):  
Open Source ◽  
Optimal Estimation ◽  
Calculation Model ◽  
The Body ◽  
Elastic Band ◽  
Visual Positioning ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
Random Sample Consensus ◽  
Pose Graph Optimization

The autonomous positioning of tunneling equipment is the key to intellectualization and robotization of a tunneling face. In this paper, a method based on simultaneous localization and mapping (SLAM) to estimate the body pose of a roadheader and build a navigation map of a roadway is presented. In terms of pose estimation, an RGB-D camera is used to collect images, and a pose calculation model of a roadheader is established based on random sample consensus (RANSAC) and iterative closest point (ICP); constructing a pose graph optimization model with closed-loop constraints. An iterative equation based on Levenberg–Marquadt is derived-, which can achieve the optimal estimation of the body pose. In terms of mapping, LiDAR is used to experimentally construct the grid map based on open-source algorithms, such as Gmapping, Cartographer, Karto, and Hector. A point cloud map, octree map, and compound map are experimentally constructed based on the open-source library RTAB-MAP. By setting parameters, such as the expansion radius of an obstacle and the updating frequency of the map, a cost map for the navigation of a roadheader is established. Combined with algorithms, such as Dijskra and timed-elastic-band, simulation experiments show that the combination of octree map and cost map can support global path planning and local obstacle avoidance.

A Framework for RF-Visual SLAM for Cooperative Multi-Agent System

2015 ◽  
pp. 697-727
Author(s):  
Herdawatie Abdul Kadir ◽  
Mohd Rizal Arshad
Keyword(s):  
Mobile Agent ◽  
Signal Strength ◽  
Received Signal Strength ◽  
Local Information ◽  
Reliable Data ◽  
Working Environment ◽  
Visual Slam ◽  
Received Signal Strength Indicator ◽  
Localization And Mapping ◽  
Multi Agent

This chapter provides a framework for radio frequency visual simultaneous localization and mapping problems for a team of agents consisting of three blimps and beacons. In a cooperative system, each agent must establish reliable data sharing during a mission. Under these conditions, a framework was proposed which allows each agent to share the local information using peer-to-peer networking schemes. The RF-vSLAM algorithm seeks to acquire a map during navigation, simultaneously localizing itself using the map and received signal strength indicator information to predict the distance between agents. In this chapter, the authors address the problem of detection features using SIFT algorithms. The authors have considered the sea surface as the working environment. In this research, the framework consisted of two types of agents, where beacon representing the static agent and blimp representing the homogeneous mobile agent. The communication exchange between these two types of agents is an environmentally friendly monitoring system that preserves natural value of the selected area.

scholarly journals A Trajectory-Based Approach to Multi-Session Underwater Visual SLAM Using Global Image Signatures

2019 ◽  
Vol 7 (8) ◽  
pp. 278 ◽  
Author(s):  
Antoni Burguera Burguera ◽  
Francisco Bonin-Font
Keyword(s):  
Feature Matching ◽  
Computational Cost ◽  
Visual Slam ◽  
Single Session ◽  
Search Region ◽  
Loop Detector ◽  
Localization And Mapping ◽  
Single Link ◽  
Random Sample Consensus ◽  
Visual Odometer

This paper presents a multi-session monocular Simultaneous Localization and Mapping (SLAM) approach focused on underwater environments. The system is composed of three main blocks: a visual odometer, a loop detector, and an optimizer. Single session loop closings are found by means of feature matching and Random Sample Consensus (RANSAC) within a search region. Multi-session loop closings are found by comparing hash-based global image signatures. The optimizer refines the trajectories and joins the different maps. Map joining preserves the trajectory structure by adding a single link between the joined sessions, making it possible to aggregate or disaggregate sessions whenever is necessary. All the optimization processes can be delayed until a certain number of loops has been found in order to reduce the computational cost. Experiments conducted in real subsea scenarios show the quality and robustness of this proposal.

scholarly journals Spare Pose Graph Decomposition and Optimization for SLAM

2019 ◽  
Vol 256 ◽  
pp. 05003
Author(s):  
Tian Liu ◽  
Yongfu Chen ◽  
Zhiyong Jin ◽  
Kai Li ◽  
Zhenting Wang ◽  
...  
Keyword(s):  
Large Scale ◽  
Optimization Method ◽  
Single Chain ◽  
Original Graph ◽  
Time Consumption ◽  
Parallel Chain ◽  
Localization And Mapping ◽  
Graph Optimization ◽  
The One ◽  
Time Resource

The graph optimization has become the mainstream technology to solve the problems of SLAM (simultaneous localization and mapping). The pose graph in the graph based SLAM is consisted with a series of nodes and edges that connect the adjacent or related poses. With the widespread use of mobile robots, the scale of pose graph has rapidly increased. Therefore, optimizing a large-scale pose graph is the bottleneck of application of graph based SLAM. In this paper, we propose an optimization method basing on the decomposition of pose graph, of which we have noticed the sparsity. With the extraction of the Single-chain and the Parallel-chain, the pose graph is decomposed into many small subgraphs. Compared with directly processing the original graph, the speed of calculation is accelerated by separately optimizing the subgraph, which is because the computational complexity is increasing exponentially with the increase of the graph’s scale. This method we proposed is very suitable for the current multi-threaded framework adopted in the mainstream SLAM, which separately calculate the subgraph decomposed by our method, rather than the original optimization requiring a large block of time in once may cause CPU obstruction. At the end of the paper, our algorithm is validated with the open source dataset of the mobile robot, of which the result illustrates our algorithm can reduce the one-time resource consumption and the time consumption of the calculation with the same map-constructing accuracy.

scholarly journals Developing a real time navigation for the mobile robots at unknown environments

2020 ◽  
Vol 20 (1) ◽  
pp. 500
Author(s):  
Sarah Haider Abdulredah ◽  
Dheyaa Jasim Kadhim
Keyword(s):  
Mobile Robot ◽  
Mean Square ◽  
Matlab Simulation ◽  
Robot Operating System ◽  
Unknown Environments ◽  
Localization And Mapping ◽  
Software Platforms ◽  
The Mean ◽  
Simulation Results ◽  
Robot Platform

<p><span>This research deals with the feasibility of a mobile robot to navigate and discover its location at unknown environments, and then constructing maps of these navigated environments for future usage. In this work, we proposed a modified Extended Kalman Filter- Simultaneous Localization and Mapping (EKF-SLAM) technique which was implemented for different unknown environments containing a different number of landmarks. Then, the detectable landmarks will play an important role in controlling the overall navigation process and EKF-SLAM technique’s performance. MATLAB simulation results of the EKF-SLAM technique come with better performance as compared with an odometry approach performance in terms of measuring the mean square error, especially when increasing the number of landmarks. After that, we simulate and evaluate a mobile robot platform named TurtleBot2e in Gazebo simulator software to achieve the using of the SLAM technique for a different environment using the Rviz library which was built on Robot Operating System in Linux. The main conclusion comes with this work is the simulation and implementation of the SLAM technique using two software platforms separately (MATLAB and ROS) in different unknown environments containing a different number of landmarks so a few number of landmark will make the mobile robot loses its path.</span></p>

scholarly journals Deep Learning for Visual SLAM in Transportation Robotics: A review

2019 ◽  
Vol 1 (3) ◽  
pp. 177-184
Author(s):  
Chao Duan ◽  
Steffen Junginger ◽  
Jiahao Huang ◽  
Kairong Jin ◽  
Kerstin Thurow
Keyword(s):  
Computer Vision ◽  
Deep Learning ◽  
Future Development ◽  
Research Progress ◽  
Visual Slam ◽  
Learning Methods ◽  
Research Results ◽  
The Past ◽  
Localization And Mapping ◽  
Challenging Environment

Abstract Visual SLAM (Simultaneously Localization and Mapping) is a solution to achieve localization and mapping of robots simultaneously. Significant achievements have been made during the past decades, geography-based methods are becoming more and more successful in dealing with static environments. However, they still cannot handle a challenging environment. With the great achievements of deep learning methods in the field of computer vision, there is a trend of applying deep learning methods to visual SLAM. In this paper, the latest research progress of deep learning applied to the field of visual SLAM is reviewed. The outstanding research results of deep learning visual odometry and deep learning loop closure detect are summarized. Finally, future development directions of visual SLAM based on deep learning is prospected.

Sign in / Sign up

Export Citation Format

Share Document