scholarly journals ON CONSTRUCTION OF A RELIABLE GROUND TRUTH FOR EVALUATION OF VISUAL SLAM ALGORITHMS

2016 ◽  
Vol 6 ◽  
pp. 1-5
Author(s):  
Jan Bayer ◽  
Petr Čížek ◽  
Jan Faigl
Keyword(s):  
Ground Truth ◽  
Accuracy Evaluation ◽  
Algorithm Performance ◽  
Localization Accuracy ◽  
Trajectory Error ◽  
Localization System ◽  
Localization And Mapping ◽  
On Line ◽  
Slam Algorithm ◽  
Frame Processing

In this work we are concerning the problem of localization accuracy evaluation of visual-based Simultaneous Localization and Mapping (SLAM) techniques. Quantitative evaluation of the SLAM algorithm performance is usually done using the established metrics of Relative pose error and Absolute trajectory error which require a precise and reliable ground truth. Such a ground truth is usually hard to obtain, while it requires an expensive external localization system. In this work we are proposing to use the SLAM algorithm itself to construct a reliable ground-truth by offline frame-by-frame processing. The generated ground-truth is suitable for evaluation of different SLAM systems, as well as for tuning the parametrization of the on-line SLAM. The presented practical experimental results indicate the feasibility of the proposed approach.

Comparison of different SLAM approaches for a driverless race car

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Nick Le Large ◽  
Frank Bieder ◽  
Martin Lauer
Keyword(s):  
Kalman Filter ◽  
Extended Kalman Filter ◽  
Ground Truth ◽  
Time Constraints ◽  
Differential Gps ◽  
Real World Data ◽  
Race Car ◽  
Localization And Mapping ◽  
Slam Algorithm ◽  
Computational Resources

Abstract For the application of an automated, driverless race car, we aim to assure high map and localization quality for successful driving on previously unknown, narrow race tracks. To achieve this goal, it is essential to choose an algorithm that fulfills the requirements in terms of accuracy, computational resources and run time. We propose both a filter-based and a smoothing-based Simultaneous Localization and Mapping (SLAM) algorithm and evaluate them using real-world data collected by a Formula Student Driverless race car. The accuracy is measured by comparing the SLAM-generated map to a ground truth map which was acquired using high-precision Differential GPS (DGPS) measurements. The results of the evaluation show that both algorithms meet required time constraints thanks to a parallelized architecture, with GraphSLAM draining the computational resources much faster than Extended Kalman Filter (EKF) SLAM. However, the analysis of the maps generated by the algorithms shows that GraphSLAM outperforms EKF SLAM in terms of accuracy.

scholarly journals Computationally Efficient Cooperative Dynamic Range-Only SLAM Based on Sum of Gaussian Filter

Sensors ◽  
2020 ◽  
Vol 20 (11) ◽  
pp. 3306
Author(s):  
Jung-Hee Kim ◽  
Doik Kim
Keyword(s):  
Convergence Rate ◽  
Computational Analysis ◽  
Dynamic Range ◽  
Estimation Error ◽  
Dynamic Environments ◽  
Map Estimation ◽  
Computationally Efficient ◽  
Localization Accuracy ◽  
Localization And Mapping ◽  
Slam Algorithm

A cooperative dynamic range-only simultaneous localization and mapping (CDRO-SLAM) algorithm based on the sum of Gaussian (SoG) filter was recently introduced. The main characteristics of the CDRO-SLAM are (i) the integration of inter-node ranges as well as usual direct robot-node ranges to improve the convergence rate and localization accuracy and (ii) the tracking of any moving nodes under dynamic environments by resetting and updating the SoG variables. In this paper, an efficient implementation of the CDRO-SLAM (eCDRO-SLAM) is proposed to mitigate the high computational burden of the CDRO-SLAM due to the inter-node measurements. Furthermore, a thorough computational analysis is presented, which reveals that the computational efficiency of the eCDRO-SLAM is significantly improved over the CDRO-SLAM. The performance of the proposed eCDRO-SLAM is compared with those of several conventional RO-SLAM algorithms and the results show that the proposed efficient algorithm has a faster convergence rate and a similar map estimation error regardless of the map size. Accordingly, the proposed eCDRO-SLAM can be utilized in various RO-SLAM applications.

Lampposts as Landmarks for Simultaneous Localization and Mapping

2011 ◽  
Vol 403-408 ◽  
pp. 823-829
Author(s):  
Musfira Jilani ◽  
Padraig Corcoran ◽  
Peter Mooney
Keyword(s):  
High Frequency ◽  
Simultaneous Localization And Mapping ◽  
Ground Truth ◽  
University Campus ◽  
Binary Classifier ◽  
Icp Algorithm ◽  
Localization And Mapping ◽  
Three Stages ◽  
Spatial Size ◽  
Slam Algorithm

This paper investigates the effectiveness of using lampposts, which are commonly found in University campus environments with high frequency, as landmarks in a 2D LIDAR based Simultaneous Localization and Mapping (SLAM) framework. Lampposts offer a number of benefits compared to other forms of landmarks. Their unique spatial signature makes it possible to design effective algorithms to extract them. They have a very small spatial size. Their use removes the challenge of determining a corresponding location between difference views. This represents a major challenge if larger objects are used as landmarks. The proposed SLAM algorithm contains three stages. Firstly LIDAR segmentation is performed. Next each object is input to a binary classifier which determines objects with a high probability of corresponding to lampposts. Finally these extracted lampposts are input to an Iterative Closest Point (ICP) based SLAM algorithm. The ICP algorithm used is an extension of the traditional ICP algorithm and filters associations due to noise. Results achieved by the proposed system were very positive. An accurate map of a university’s lampposts was created and localization, when compared to GPS ground-truth, was very accurate.

scholarly journals A Versatile Decentralized 3D Volumetric Fusion for On-line Reconstruction

2020 ◽  
Vol 10 (6) ◽  
pp. 6584-6588
Author(s):  
A. Rajput ◽  
A. Hussain ◽  
F. Akhtar ◽  
Z. H. Khand ◽  
H. Magsi
Keyword(s):  
3D Reconstruction ◽  
3D Models ◽  
Depth Sensing ◽  
Computing Device ◽  
Depth Data ◽  
Resource Limited ◽  
Sensing Technology ◽  
Localization And Mapping ◽  
On Line ◽  
Slam Algorithm

Advancement in depth-sensing technology has allowed mobile robots to visualize the surrounding environment in 3D models. Regardless of the sensing technology (i.e. active, passive, or laser-based), a complete system that integrates recent depth data in previous 3D models in real-time is done by employing Simultaneous Localization And Mapping (SLAM) algorithms followed by a 3D reconstruction engine. Unfortunately, both the SLAM algorithm and the 3D reconstruction engine are usually executed on a single computing device, making the whole system exceptionally costly and heavy and restricting the robot's mobility. This paper proposes a decentralized, modular reconstruction system capable of employing various sensors to facilitate online 3D reconstruction from a resource-limited mobile robot.

scholarly journals A V-SLAM Guided and Portable System for Photogrammetric Applications

2021 ◽  
Vol 13 (12) ◽  
pp. 2351
Author(s):  
Alessandro Torresani ◽  
Fabio Menna ◽  
Roberto Battisti ◽  
Fabio Remondino
Keyword(s):  
Low Cost ◽  
Cost Effective ◽  
Motion Blur ◽  
Guidance System ◽  
Prototype System ◽  
Active Sensors ◽  
Localization And Mapping ◽  
Acquisition Speed ◽  
Mobile Mapping System ◽  
Slam Algorithm

Mobile and handheld mapping systems are becoming widely used nowadays as fast and cost-effective data acquisition systems for 3D reconstruction purposes. While most of the research and commercial systems are based on active sensors, solutions employing only cameras and photogrammetry are attracting more and more interest due to their significantly minor costs, size and power consumption. In this work we propose an ARM-based, low-cost and lightweight stereo vision mobile mapping system based on a Visual Simultaneous Localization And Mapping (V-SLAM) algorithm. The prototype system, named GuPho (Guided Photogrammetric System) also integrates an in-house guidance system which enables optimized image acquisitions, robust management of the cameras and feedback on positioning and acquisition speed. The presented results show the effectiveness of the developed prototype in mapping large scenarios, enabling motion blur prevention, robust camera exposure control and achieving accurate 3D results.

scholarly journals An Indoor Localization System Using Residual Learning with Channel State Information

Entropy ◽  
2021 ◽  
Vol 23 (5) ◽  
pp. 574
Author(s):  
Chendong Xu ◽  
Weigang Wang ◽  
Yunwei Zhang ◽  
Jie Qin ◽  
Shujuan Yu ◽  
...  
Keyword(s):  
Channel State Information ◽  
Indoor Environment ◽  
Indoor Localization ◽  
Location Based Services ◽  
Channel State ◽  
State Information ◽  
Localization Accuracy ◽  
Localization System ◽  
Increasing Demand ◽  
Degradation Problem

With the increasing demand of location-based services, neural network (NN)-based intelligent indoor localization has attracted great interest due to its high localization accuracy. However, deep NNs are usually affected by degradation and gradient vanishing. To fill this gap, we propose a novel indoor localization system, including denoising NN and residual network (ResNet), to predict the location of moving object by the channel state information (CSI). In the ResNet, to prevent overfitting, we replace all the residual blocks by the stochastic residual blocks. Specially, we explore the long-range stochastic shortcut connection (LRSSC) to solve the degradation problem and gradient vanishing. To obtain a large receptive field without losing information, we leverage the dilated convolution at the rear of the ResNet. Experimental results are presented to confirm that our system outperforms state-of-the-art methods in a representative indoor environment.

scholarly journals An Improved Localization Method for the Transition between Autonomous Underwater Vehicle Homing and Docking

Sensors ◽  
2021 ◽  
Vol 21 (7) ◽  
pp. 2468
Author(s):  
Ri Lin ◽  
Feng Zhang ◽  
Dejun Li ◽  
Mingwei Lin ◽  
Gengli Zhou ◽  
...  
Keyword(s):  
Data Exchange ◽  
Autonomous Underwater Vehicle ◽  
Autonomous Underwater Vehicles ◽  
Optimization Method ◽  
Short Baseline ◽  
Localization Accuracy ◽  
Localization Method ◽  
Mapping Algorithm ◽  
Localization And Mapping ◽  
Update Rate

Docking technology for autonomous underwater vehicles (AUVs) involves energy supply, data exchange and navigation, and plays an important role to extend the endurance of the AUVs. The navigation method used in the transition between AUV homing and docking influences subsequent tasks. How to improve the accuracy of the navigation in this stage is important. However, when using ultra-short baseline (USBL), outliers and slow localization updating rates could possibly cause localization errors. Optical navigation methods using underwater lights and cameras are easily affected by the ambient light. All these may reduce the rate of successful docking. In this paper, research on an improved localization method based on multi-sensor information fusion is carried out. To improve the localization performance of AUVs under motion mutation and light variation conditions, an improved underwater simultaneous localization and mapping algorithm based on ORB features (IU-ORBSALM) is proposed. A nonlinear optimization method is proposed to optimize the scale of monocular visual odometry in IU-ORBSLAM and the AUV pose. Localization tests and five docking missions are executed in a swimming pool. The localization results indicate that the localization accuracy and update rate are both improved. The 100% successful docking rate achieved verifies the feasibility of the proposed localization method.

scholarly journals Multi-Focal Vision and Gaze Control Improve Navigation Performance

10.5772/50920 ◽  
2012 ◽  
Vol 9 (1) ◽  
pp. 25 ◽  
Author(s):  
Kolja Kühnlenz ◽  
Martin Buss
Keyword(s):  
Mobile Robot ◽  
Robot Navigation ◽  
Active Vision ◽  
Path Following ◽  
Mobile Robot Navigation ◽  
Vision Systems ◽  
Localization Accuracy ◽  
Localization And Mapping ◽  
Novel Concept ◽  
The Impact

Multi-focal vision systems comprise cameras with various fields of view and measurement accuracies. This article presents a multi-focal approach to localization and mapping of mobile robots with active vision. An implementation of the novel concept is done considering a humanoid robot navigation scenario where the robot is visually guided through a structured environment with several landmarks. Various embodiments of multi-focal vision systems are investigated and the impact on navigation performance is evaluated in comparison to a conventional mono-focal stereo set-up. The comparative studies clearly show the benefits of multi-focal vision for mobile robot navigation: flexibility to assign the different available sensors optimally in each situation, enhancement of the visible field, higher localization accuracy, and, thus, better task performance, i.e. path following behavior of the mobile robot. It is shown that multi-focal vision may strongly improve navigation performance.

scholarly journals DSCP: Depthwise Separable Convolution-Based Passive Indoor Localization Using CSI Fingerprint

2021 ◽  
Vol 2021 ◽  
pp. 1-17
Author(s):  
Chong Han ◽  
Wenjing Xun ◽  
Lijuan Sun ◽  
Zhaoxiao Lin ◽  
Jian Guo
Keyword(s):  
Indoor Localization ◽  
Wireless Sensing ◽  
Training Phase ◽  
Training Time ◽  
Localization Accuracy ◽  
Localization System ◽  
Complex Models ◽  
Indoor Scenarios ◽  
Trained Network ◽  
Localization Systems

Wi-Fi-based indoor localization has received extensive attention in wireless sensing. However, most Wi-Fi-based indoor localization systems have complex models and high localization delays, which limit the universality of these localization methods. To solve these problems, a depthwise separable convolution-based passive indoor localization system (DSCP) is proposed. DSCP is a lightweight fingerprint-based localization system that includes an offline training phase and an online localization phase. In the offline training phase, the indoor scenario is first divided into different areas to set training locations for collecting CSI. Then, the amplitude differences of these CSI subcarriers are extracted to construct location fingerprints, thereby training the convolutional neural network (CNN). In the online localization phase, CSI data are first collected at the test locations, and then, the location fingerprint is extracted and finally fed to the trained network to obtain the predicted location. The experimental results show that DSCP has a short training time and a low localization delay. DSCP achieves a high localization accuracy, above 97%, and a small median localization distance error of 0.69 m in typical indoor scenarios.

scholarly journals FDMine: a graph mining approach to predict and evaluate food-drug interactions

2021 ◽  
Author(s):  
Md Mostafizur Rahman ◽  
Srinivas Mukund Vadrev ◽  
Arturo Magana-Mora ◽  
Jacob Levman ◽  
Othman Soufan
Keyword(s):  
Drug Interactions ◽  
Graph Mining ◽  
Characteristic Curve ◽  
Ground Truth ◽  
Recovery Process ◽  
Food Composition ◽  
Accuracy Evaluation ◽  
Chemical Structures ◽  
Food Items ◽  
Drug Similarity

Abstract Food-drug interactions (FDIs) arise when nutritional dietary consumption regulates biochemical mechanisms involved in drug metabolism. These interactions can create unexpected adverse pharmacological effects. By contrast, particular foods can aid in the recovery process of a patient. Towards characterizing the nature of food’s influence on pharmacological treatment, it is essential to detect all possible FDIs. In this study, we propose FDMine, a novel systematic framework that models the FDI problem as a homogenous graph. In this graph, all nodes representing drug, food and food composition are referenced as chemical structures. This homogenous representation enables us to take advantage of reported drug-drug interactions for accuracy evaluation, especially when accessible ground truth for FDIs is lacking. Our dataset consists of 788 unique approved small molecule drugs with metabolism-related drug-drug interactions (DDIs) and 320 unique food items, composed of 563 unique compounds with 179 health effects. The potential number of interactions is 87,192 and 92,143 when two different versions of the graph referred to as disjoint and joint graphs are considered, respectively. We defined several similarity subnetworks comprising food-drug similarity (FDS), drug-drug similarity (DDS), and food-food similarity (FFS) networks, based on similarity profiles. A unique part of the graph is the encoding of the food composition as a set of nodes and calculating a content contribution score to re-weight the similarity links. To predict new FDI links, we applied the path category-based (path length 2 and 3) and neighborhood-based similarity-based link prediction algorithms. We calculated the precision@top (top 1%, 2%, and 5%) of the newly predicted links, the area under the receiver operating characteristic curve, and precision-recall curve. We have performed three types of evaluations to benchmark results using different types of interactions. The shortest path-based method has achieved a precision 84%, 60% and 40% for the top 1%, 2% and 5% of FDIs identified, respectively. We validated the top FDIs predicted using FDMine to demonstrate its applicability and we relate therapeutic anti-inflammatory effects of food items informed by FDIs. We hypothesize that the proposed framework can be used to gain new insights on FDIs. FDMine is publicly available to support clinicians and researchers.

Sign in / Sign up

Export Citation Format

Share Document