An online multi-robot SLAM system for 3D LiDARs

Dubé, Renaud; Gawel, Abel; Sommer, Hannes; Nieto, Juan; Siegwart, Roland; Cadena, César

doi:10.1109/iros.2017.8202268

Cited by 117 publications

(72 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The results are indicative as the ICP setting of [7] slightly differs from the setting of [9] we use; odo.+ICP : combines odometry N R (T odo , Q odo ) and ICP N R (T icp ,Q icp ) measurements with our proposed covariance estimates considering them as independent measurements, as in Figure 2 (left) and e.g. [2]; proposed : involves odometry N R (T odo , Q odo ) and ICP N R (T icp ,Q icp ) estimates along with the crosscorrelation termQ cross , see Figure 2 (right). Based on Section II, we prune ICP registration from the posegraph using a Neyman-Pearson statistical test about difference between the ICP and odometry log(T −1 icpT odo ).…”

Section: A Compared Methods and Evaluation Metricsmentioning

confidence: 99%

“…2. ICP computes rigid transformationT icp that may serve as a laser odometry sensor which is combined with other types of odometry based on inertial sensors and/or differential wheel speeds, that provide relative transformationT odo [2,3]. To perform fusion for localization one may use, e.g., pose-graph optimization like GTSAM [23] or g 2 o [24], that requires evaluating odometry uncertainty Q odo and ICP uncertainty Q icp .…”

Section: Contributions and Paper's Organizationmentioning

confidence: 99%

See 1 more Smart Citation

A New Approach to 3D ICP Covariance Estimation

Brossard

Bonnabel

Barrau

2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

In mobile robotics, scan matching of point clouds using Iterative Closest Point (ICP) allows estimating sensor displacements. It may prove important to assess the associated uncertainty about the obtained rigid transformation, especially for sensor fusion purposes. In this paper we propose a novel approach to 3D ICP covariance computation that accounts for all the sources of errors as listed in Censi's pioneering work, namely wrong convergence, underconstrained situations, and sensor noise. Our approach builds on two facts. First, ICP is not a standard sensor: owing to wrong convergence the concept of ICP covariance per se is actually meaningless, as the dispersion in the ICP outputs may largely depend on the accuracy of the initialization, and is thus inherently related to the prior uncertainty on the displacement. We capture this using the unscented transform, which also reflects correlations between initial and final uncertainties. Then, assuming white sensor noise leads to overoptimism: ICP is biased, owing to e.g. calibration biases, which we account for. Our solution is tested on publicly available real data ranging from structured to unstructured environments, where our algorithm predicts consistent results with actual uncertainty, and compares very favorably to previous methods. We finally demonstrate the benefits of our method for pose-graph localization, where our approach improves accuracy and robustness of the estimates.

show abstract

Section: A Compared Methods and Evaluation Metricsmentioning

confidence: 99%

Section: Contributions and Paper's Organizationmentioning

confidence: 99%

A New Approach to 3D ICP Covariance Estimation

Brossard

Bonnabel

Barrau

2020

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…WINS achieves decimeter-level positioning accuracy. It is not designed to defeat the accuracies of CV/laser based approaches [6], [19], [20], which can achieve centimeter-level when environments are suitable, e.g., well-lighted, texture-rich, clear in line of sight (LOS). But our design complements them to support navigation in a lightweight and low-cost manner and we are highly resilient to LOS limitations so as to work in vision/lasercrippled scenarios.…”

Section: F Limitationsmentioning

confidence: 99%

Wi-Fi-Inertial Indoor Pose Estimation for Microaerial Vehicles

Zhang

Wang

Jiang

2021

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

This paper presents an indoor pose estimation system for micro aerial vehicles (MAVs) with a single WiFi access point. Conventional approaches based on computer vision are limited by illumination conditions and environmental texture. Our system is free of visual limitations and instantly deployable, working upon existing WiFi infrastructure without any deployment cost. Our system consists of two coupled modules. First, we propose an angle-of-arrival (AoA) estimation algorithm to estimate MAV attitudes and disentangle the AoA for positioning. Second, we formulate a WiFi-inertial sensor fusion model that fuses the AoA and the odometry measured by inertial sensors to optimize MAV poses. Considering the practicality of MAVs, our system is designed to be real-time and initializationfree for the need of agile flight in unknown environments. The indoor experiments show that our system achieves the accuracy of pose estimation with the position error of 61.7 cm and the attitude error of 0.92°. Index Terms-Micro aerial vehicle, pose estimation, WiFi.

show abstract

“…One might argue that in large buildings for example airport, not all users share the common place. It is possible to first build several sub-maps, and then merge them into a large and consistent map [51] [52] [53]. An edge is added to connect the first nodes in different tracks.…”

Section: F Merging Tracks At Different Timesmentioning

confidence: 99%

Collaborative SLAM Based on WiFi Fingerprint Similarity and Motion Information

Liu

Marakkalage

Padmal

et al. 2020

IEEE Internet Things J.

View full text Add to dashboard Cite

Simultaneous localization and mapping (SLAM) has been extensively researched in past years particularly with regard to range-based or visual-based sensors. Instead of deploying dedicated devices that use visual features, it is more pragmatic to exploit the radio features to achieve this task, due to their ubiquitous nature and the widespread deployment of Wi-Fi wireless network. This paper presents a novel approach for collaborative simultaneous localization and radio fingerprint mapping (C-SLAM-RF) in large unknown indoor environments. The proposed system uses received signal strengths (RSS) from Wi-Fi access points (AP) in the existing infrastructure and pedestrian dead reckoning (PDR) from a smart phone, without a prior knowledge about map or distribution of AP in the environment. We claim a loop closure based on the similarity of the two radio fingerprints. To further improve the performance, we incorporate the turning motion and assign a small uncertainty value to a loop closure if a matched turning is identified. The experiment was done in an area of 130 meters by 70 meters and the results show that our proposed system is capable of estimating the tracks of four users with an accuracy of 0.6 meters with Tango-based PDR and 4.76 meters with a step counter-based PDR.

show abstract

An online multi-robot SLAM system for 3D LiDARs

Cited by 117 publications

References 25 publications

A New Approach to 3D ICP Covariance Estimation

A New Approach to 3D ICP Covariance Estimation

Wi-Fi-Inertial Indoor Pose Estimation for Microaerial Vehicles

Collaborative SLAM Based on WiFi Fingerprint Similarity and Motion Information

Contact Info

Product

Resources

About