Fast and Effective Loop Closure Detection to Improve SLAM Performance

Guclu, Oguzhan; Can, Ahmet Burak

doi:10.1007/s10846-017-0718-z

Cited by 26 publications

(9 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A feature-only map is obtained from a decoupled SLAM (D-SLAM) and considers all landmarks features [Zhao et al, 2019]. Building an accurate map, along with performing precise localization of robots is a non easy problem [Guclu and Can, 2019], that demands a variety of information. Thus, when a developer or researcher implements a solution for the SLAM problem, it is important to know what information must be stored.…”

Section: Slam Principles: Preliminariesmentioning

confidence: 99%

A categorization of simultaneous localization and mapping knowledge for mobile robots

Lupa

Ticona-Herrera

Cardinale

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

Los robots autónomos están desempeñando un papel importante en las actividades académicas, tecnológicas y científicas. Por lo tanto, su comportamiento se está volviendo más complejo. Las principales tareas de los robots autónomos incluyen el mapeo de un entorno y la localización de sí mismos. Estas tareas comprenden el problema de la Localización y Mapeo Simultáneo (SLAM). La representación del conocimiento SLAM (por ejemplo, las características de los robots, la información del medio ambiente, el mapeo y la información de localización), con un modelo estándar y bien definido, proporciona la base para desarrollar soluciones eficientes e interoperables. Sin embargo, hasta donde sabemos, no existe una clasificación común de esos conocimientos. Muchos trabajos existentes basados en la Web Semántica, han formulado ontologías para modelar información relacionada sólo con algunos aspectos del SLAM, sin un estándar. En este trabajo, proponemos una categorización del conocimiento manejado en el SLAM, basado en las ontolog´ıas existentes y los principios delSLAM. También clasificamos ontologías recientes y populares de acuerdo a las categor´ıas propuestas y resaltamos las lecciones a aprender de las ontolog´ıas existentes. Evidenciando la necesidad de desarrollar una ontolog´ıa completa para representar la información de SLAM en los robot móviles.

show abstract

Section: Slam Principles: Preliminariesmentioning

confidence: 99%

A categorization of simultaneous localization and mapping knowledge for mobile robots

Lupa

Ticona-Herrera

Cardinale

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

View full text Add to dashboard Cite

show abstract

“…It was resulting in increased accuracy of the algorithm and reduced mathematical processing burden. It is also possible to find works on sub-mapping and loop-closure problems in the literature [27].…”

Section: Literature Reviewmentioning

confidence: 99%

Exact flow of particles using for state estimations in unmanned aerial systems` navigation

2020

View full text Add to dashboard Cite

The navigation is a substantial issue in the field of robotics. Simultaneous Localization and Mapping (SLAM) is a principle for many autonomous navigation applications, particularly in the Global Navigation Satellite System (GNSS) denied environments. Many SLAM methods made substantial contributions to improve its accuracy, cost, and efficiency. Still, it is a considerable challenge to manage robust SLAM, and there exist several attempts to find better estimation algorithms for it. In this research, we proposed a novel Bayesian filtering based Airborne SLAM structure for the first time in the literature. We also presented the mathematical background of the algorithm, and the SLAM model of an autonomous aerial vehicle. Simulation results emphasize that the new Airborne SLAM performance with the exact flow of particles using for recursive state estimations superior to other approaches emerged before, in terms of accuracy and speed of convergence. Nevertheless, its computational complexity may cause real-time application concerns, particularly in high-dimensional state spaces. However, in Airborne SLAM, it can be preferred in the measurement environments that use low uncertainty sensors because it gives more successful results by eliminating the problem of degeneration seen in the particle filter structure.

show abstract

“…The first step is searching candidate keyframes (visual place recognition) and the second step is keypoint feature matching (metric localization). Without accurate visual place recognition, trajectory drift will occur and an ambiguous map of unknown environment will be constructed in large-scale localization and mapping [3,4]. However, visual place recognition remains a challenge problem because of perceptual aliasing and perceptual variability problems.…”

Section: Introductionmentioning

confidence: 99%

Towards a Robust Visual Place Recognition in Large-Scale vSLAM Scenarios Based on a Deep Distance Learning

Chen

Jin

Xia

2021

Sensors

View full text Add to dashboard Cite

The application of deep learning is blooming in the field of visual place recognition, which plays a critical role in visual Simultaneous Localization and Mapping (vSLAM) applications. The use of convolutional neural networks (CNNs) achieve better performance than handcrafted feature descriptors. However, visual place recognition is still a challenging task due to two major problems, i.e., perceptual aliasing and perceptual variability. Therefore, designing a customized distance learning method to express the intrinsic distance constraints in the large-scale vSLAM scenarios is of great importance. Traditional deep distance learning methods usually use the triplet loss which requires the mining of anchor images. This may, however, result in very tedious inefficient training and anomalous distance relationships. In this paper, a novel deep distance learning framework for visual place recognition is proposed. Through in-depth analysis of the multiple constraints of the distance relationship in the visual place recognition problem, the multi-constraint loss function is proposed to optimize the distance constraint relationships in the Euclidean space. The new framework can support any kind of CNN such as AlexNet, VGGNet and other user-defined networks to extract more distinguishing features. We have compared the results with the traditional deep distance learning method, and the results show that the proposed method can improve the performance by 19–28%. Additionally, compared to some contemporary visual place recognition techniques, the proposed method can improve the performance by 40%/36% and 27%/24% in average on VGGNet/AlexNet using the New College and the TUM datasets, respectively. It’s verified the method is capable to handle appearance changes in complex environments.

show abstract

Fast and Effective Loop Closure Detection to Improve SLAM Performance

Cited by 26 publications

References 34 publications

A categorization of simultaneous localization and mapping knowledge for mobile robots

A categorization of simultaneous localization and mapping knowledge for mobile robots

Exact flow of particles using for state estimations in unmanned aerial systems` navigation

Towards a Robust Visual Place Recognition in Large-Scale vSLAM Scenarios Based on a Deep Distance Learning

Contact Info

Product

Resources

About