Collar Line Segments for fast odometry estimation from Velodyne point clouds

Veľas, Martin; Španěl, Michal; Herout, Adam

doi:10.1109/icra.2016.7487648

Cited by 85 publications

(71 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When using Velodyne LiDAR data, the authors achieved ±20 cm accuracy in the registration of pairwise scans. In our evaluation [10] using KITTI dataset [3], the method yields average error 11.5cm in the frameto-frame registration task. The robustness of GICP drops in case of large distance between the scans (> 6m).…”

Section: Related Workmentioning

confidence: 99%

CNN for IMU assisted odometry estimation using velodyne LiDAR

Veľas

Španěl

Hradiš

et al. 2018

2018 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)

Self Cite

View full text Add to dashboard Cite

We introduce a novel method for odometry estimation using convolutional neural networks from 3D LiDAR scans. The original sparse data are encoded into 2D matrices for the training of proposed networks and for the prediction. Our networks show significantly better precision in the estimation of translational motion parameters comparing with state of the art method LOAM, while achieving real-time performance. Together with IMU support, high quality odometry estimation and LiDAR data registration is realized. Moreover, we propose alternative CNNs trained for the prediction of rotational motion parameters while achieving results also comparable with state of the art. The proposed method can replace wheel encoders in odometry estimation or supplement missing GPS data, when the GNSS signal absents (e.g. during the indoor mapping). Our solution brings real-time performance and precision which are useful to provide online preview of the mapping results and verification of the map completeness in real time.

show abstract

Section: Related Workmentioning

confidence: 99%

CNN for IMU assisted odometry estimation using velodyne LiDAR

Veľas

Španěl

Hradiš

et al. 2018

2018 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Laser-based odometry and mapping systems often reduce the 3D point cloud data by relying on features [34], subsampled clouds [16,30], or voxelbased [34] as well as NDT-based map representations [27,23,4]. In contrast to that, we operate on all laser points and perform a registration to a surfel map at every step of the algorithm.…”

Section: Related Workmentioning

confidence: 99%

“…Most of these approaches use (semi-)dense reconstructions of the environment and exploit them for frameto-model tracking, either by jointly optimizing the map and pose estimates or by alternating pose estimation and map building [21]. Dense approaches have a prospective advantage over feature-based and sparse approaches as they use all available information and thus do not depend on reliable feature extraction or availability of such features.In contrast to these developments, current 3D laser-based mapping systems mainly accomplish the estimation relying on feature-based solutions [34,35], reduced map representations [14,13], voxel grid-based methods [16], or point sub-sampling [30], which all effectively reduce the data used for alignment. Compared to most indoor applications using RGB-D sensors, we have to tackle additional challenges in outdoor applications using 3D laser sensors, i.e., (1) fast sensor movement resulting in large displacements between scans, (2) comparably sparse point clouds, and (3) large-scale environments.…”

mentioning

confidence: 99%

“…In contrast to these developments, current 3D laser-based mapping systems mainly accomplish the estimation relying on feature-based solutions [34,35], reduced map representations [14,13], voxel grid-based methods [16], or point sub-sampling [30], which all effectively reduce the data used for alignment. Compared to most indoor applications using RGB-D sensors, we have to tackle additional challenges in outdoor applications using 3D laser sensors, i.e., (1) fast sensor movement resulting in large displacements between scans, (2) comparably sparse point clouds, and (3) large-scale environments.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Efficient Surfel-Based SLAM using 3D Laser Range Data in Urban Environments

Behley¹,

Stachniss²

2018

Robotics: Science and Systems XIV

374

264

View full text Add to dashboard Cite

Abstract-Accurate and reliable localization and mapping is a fundamental building block for most autonomous robots. For this purpose, we propose a novel, dense approach to laserbased mapping that operates on three-dimensional point clouds obtained from rotating laser sensors. We construct a surfel-based map and estimate the changes in the robot's pose by exploiting the projective data association between the current scan and a rendered model view from that surfel map. For detection and verification of a loop closure, we leverage the map representation to compose a virtual view of the map before a potential loop closure, which enables a more robust detection even with low overlap between the scan and the already mapped areas. Our approach is efficient and enables real-time capable registration. At the same time, it is able to detect loop closures and to perform map updates in an online fashion. Our experiments show that we are able to estimate globally consistent maps in large scale environments solely based on point cloud data.I. INTRODUCTION Most autonomous robots, including self-driving cars, must be able to reliably localize themselves, ideally by using only their own sensors without relying on external information such as GPS or other additional infrastructure placed in the environment. There has been significant advances in visionbased [6,7] and RGB-D-based [18,33,3] SLAM systems over the past few years. Most of these approaches use (semi-)dense reconstructions of the environment and exploit them for frameto-model tracking, either by jointly optimizing the map and pose estimates or by alternating pose estimation and map building [21]. Dense approaches have a prospective advantage over feature-based and sparse approaches as they use all available information and thus do not depend on reliable feature extraction or availability of such features.In contrast to these developments, current 3D laser-based mapping systems mainly accomplish the estimation relying on feature-based solutions [34,35], reduced map representations [14,13], voxel grid-based methods [16], or point sub-sampling [30], which all effectively reduce the data used for alignment. Compared to most indoor applications using RGB-D sensors, we have to tackle additional challenges in outdoor applications using 3D laser sensors, i.e., (1) fast sensor movement resulting in large displacements between scans, (2) comparably sparse point clouds, and (3) large-scale environments.In this paper, we present a dense mapping approach called Surfel-based Mapping (SuMa), which builds globally consistent maps by tracking the pose, so-called odometry, and closing loops using a pose graph as illustrated in Figure 1. To this end, we employ a surfel map to efficiently generate synthetic views for projective data association and additional

show abstract

“…When the scanning rate is slow, a scan cannot be consider as a rigid body but distortion is present due to external motion of the laser scanner. To date, it has been shown that motion can be recovered with a laser scanner itself (Ceriani, Sanchez, Taddei, Wolfart, & Sequeira, ; Velas, Spanel, & Herout, ; Wei, Wu, & Fu, ). This requires a motion model being involved.…”

Section: Related Workmentioning

confidence: 99%

Laser–visual–inertial odometry and mapping with high robustness and low drift

Zhang¹,

Singh²

2018

Journal of Field Robotics

200

127

View full text Add to dashboard Cite

We present a data processing pipeline to online estimate ego-motion and build a map of the traversed environment, leveraging data from a 3D laser scanner, a camera, and an inertial measurement unit (IMU). Different from traditional methods that use a Kalman filter or factor-graph optimization, the proposed method employs a sequential, multilayer processing pipeline, solving for motion from coarse to fine.Starting with IMU mechanization for motion prediction, a visual-inertial coupled method estimates motion; then, a scan matching method further refines the motion estimates and registers maps. The resulting system enables high-frequency, lowlatency ego-motion estimation, along with dense, accurate 3D map registration.Further, the method is capable of handling sensor degradation by automatic reconfiguration bypassing failure modules. Therefore, it can operate in the presence of highly dynamic motion as well as in the dark, texture-less, and structure-less environments. During experiments, the method demonstrates 0.22% of relative position drift over 9.3 km of navigation and robustness w.r.t. running, jumping, and even highway speed driving (up to 33 m/s).mapping, ego-motion estimation, 3D laser scanner, vision, inertial sensor 1 | INTRODUCTION This paper aims at developing a method for online ego-motion estimation with data from a 3D laser scanner, a camera, and an inertial measurement unit (IMU). The estimated motion further registers laser points to build a map of the traversed environment. In many real-world applications, ego-motion estimation and mapping must be conducted in real time. In an autonomous navigation system, the map is crucial for motion planning and obstacle avoidance, while the motion estimation is important for vehicle control and maneuver.Very often, high-accuracy global positioning system (GPS)-inertial navigation system (INS) solutions are impractical when the application is GPS-denied, lightweight, or cost-sensitive. The proposed method utilizes only perception sensors without reliance on GPS.Specially, we are interested in solving for extremely aggressive motion. Yet, it remains nonobvious how to solve the problem reliably in 6 degrees of freedom (DOF), in real time, and in a small form factor. The problem is closely relevant to sensor degradation due to sparsity of the data during dynamic maneuver. To the best of our knowledge, the proposed method is by far the first to enable such high-rate ego-motion estimation capable of handling running and jumping (see Figure 1 as an example), while at the same time develop a dense, accurate 3D map, in the field under various lighting and structural conditions, and using only sensing and computing devices that can be easily carried by a person.The key reason that enables this level of performance is our novel way of data processing. As shown in Figure 2a, a standard Kalman filter-based method typically uses IMU mechanization in a prediction step followed by update steps seeded with individual visual features or laser landmarks. On the other hand, a fa...

show abstract

Collar Line Segments for fast odometry estimation from Velodyne point clouds

Cited by 85 publications

References 15 publications

CNN for IMU assisted odometry estimation using velodyne LiDAR

CNN for IMU assisted odometry estimation using velodyne LiDAR

Efficient Surfel-Based SLAM using 3D Laser Range Data in Urban Environments

Laser–visual–inertial odometry and mapping with high robustness and low drift

Contact Info

Product

Resources

About