LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM

Cattaneo, Daniele; Vaghi, Matteo; Valada, Abhinav

doi:10.1109/tro.2022.3150683

Cited by 131 publications

(82 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Loop Closure Generation: While there are numerous ways to generate potential loop closures, such as place recognition [20]- [22] or junction detection [6], we use proximity based generation in our experiments. This approach generates candidates from nodes that lie withing a certain distance d from the most recent node in the pose graph; d is adaptive and is defined as d = α|n current − n candidate |, which is dependent on the relative traversal between two nodes for the single-robot case and d = αn current , which is dependent on the absolute traversal for the multi-robot case, where n current and n candidate are the index of the current and candidate nodes respectively and α is a constant (0.2m).…”

Section: A Slam System Overviewmentioning

confidence: 99%

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Denniston

Chang

Reinke

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Multi-robot SLAM systems in GPS-denied environments require loop closures to maintain a drift-free centralized map. With an increasing number of robots and size of the environment, checking and computing the transformation for all the loop closure candidates becomes computationally infeasible. In this work, we describe a loop closure module that is able to prioritize which loop closures to compute based on the underlying pose graph, the proximity to known beacons, and the characteristics of the point clouds. We validate this system in the context of the DARPA Subterranean Challenge and on four challenging underground datasets where we demonstrate the ability of this system to generate and maintain a map with low error. We find that our proposed techniques are able to select effective loop closures which results in 51% mean reduction in median error when compared to an odometric solution and 75% mean reduction in median error when compared to a baseline version of this system with no prioritization. We also find our proposed system is able to achieve a lower error in the mission time of one hour when compared to a system that processes every possible loop closure in four and a half hours. The code and dataset for this work can be found at https://github.com/NeBula-Autonomy/LAMP.

show abstract

Section: A Slam System Overviewmentioning

confidence: 99%

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Denniston

Chang

Reinke

et al. 2022

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Mapping: Spatial maps built with Simultaneous Localization and Mapping (SLAM) have been used for tasks such as exploration [28,3] and FPS games [2]. Both occupancy and semantic maps have commonly been used in embodied AI tasks [25,16] and several works have presented approaches to build such maps in complex environments [4,5].…”

Section: Related Workmentioning

confidence: 99%

Learning Long-Horizon Robot Exploration Strategies for Multi-Object Search in Continuous Action Spaces

Fabian¹,

Honerkamp²,

Welschehold³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recent advances in vision-based navigation and exploration have shown impressive capabilities in photorealistic indoor environments. However, these methods still struggle with long-horizon tasks and require large amounts of data to generalize to unseen environments. In this work, we present a novel reinforcement learning approach for multi-object search that combines short-term and long-term reasoning in a single model while avoiding the complexities arising from hierarchical structures. In contrast to existing multi-object search methods that act in granular discrete action spaces, our approach achieves exceptional performance in continuous action spaces. We perform extensive experiments and show that it generalizes to unseen apartment environments with limited data. Furthermore, we demonstrate zero-shot transfer of the learned policies to an office environment in real world experiments.

show abstract

“…Loop Closure Generation: While there are numerous ways to generate potential loop closures, such as place recognition [14], [17], [18] or junction detection [4], we use proximity based generation in our experiments. This approach generates candidates from nodes that lie withing a certain distance d from the most recent node in the pose graph; d is adaptive and is defined as d = α|n current − n candidate |, which is dependent on the relative traversal between two nodes for the single-robot case and d = αn current , which is dependent on the absolute traversal for the multi-robot case, where n current and n candidate are the index of the current and candidate nodes respectively and α is a constant (0.2m in our experiments).…”

Section: A Slam System Overviewmentioning

confidence: 99%

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Denniston¹,

Chang²,

Reinke³

et al. 2022

Preprint

View full text Add to dashboard Cite

Multi-robot SLAM systems in GPS-denied environments require loop closures to maintain a drift-free centralized map. With increasing number of robots and size of the environment, checking and computing the transformation for all the loop closure candidates becomes computationally infeasible. In this work, we describe a loop closure module that is able to prioritize which loop closures to compute based on the underlying pose graph, the proximity to known beacons, and the characteristics of the point clouds. We validate this system in the context of the DARPA Subterranean Challenge and on numerous challenging underground datasets and demonstrate the ability of this system to generate and maintain a map with low error. We find that our proposed techniques are able to select effective loop closures which results in 51% mean reduction in median error when compared to an odometric solution and 75% mean reduction in median error when compared to a baseline version of this system with no prioritization. We also find our proposed system is able to find a lower error in the mission time of one hour when compared to a system which processes every possible loop closure in four and a half hours.

show abstract

LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM

Cited by 131 publications

References 53 publications

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Learning Long-Horizon Robot Exploration Strategies for Multi-Object Search in Continuous Action Spaces

Loop Closure Prioritization for Efficient and Scalable Multi-Robot SLAM

Contact Info

Product

Resources

About