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.

We present a novel way of odometry estimation from Velodyne LiDAR point cloud scans. The aim of our work is to overcome the most painful issues of Velodyne data - the sparsity and the quantity of data points - in an efficient way, enabling more precise registration. Alignment of the point clouds which yields the final odometry is based on random sampling of the clouds using Collar Line Segments (CLS). The closest line segment pairs are identified in two sets of line segments obtained from two consequent Velodyne scans. From each pair of correspondences, a transformation aligning the matched line segments into a 3D plane is estimated. By this, significant planes (ground, walls, ...) are preserved among aligned point clouds. Evaluation using the KITTI dataset shows that our method outperforms publicly available and commonly used state-of-the-art method GICP for point cloud registration in both accuracy and speed, especially in cases where the scene lacks significant landmarks or in typical urban elements. For such environments, the registration error of our method is reduced by 75% compared to the original GICP error.

This paper presents a novel method for ground segmentation in Velodyne point clouds. We propose an encoding of sparse 3D data from the Velodyne sensor suitable for training a convolutional neural network (CNN). This general purpose approach is used for segmentation of the sparse point cloud into ground and non-ground points. The LiDAR data are represented as a multi-channel 2D signal where the horizontal axis corresponds to the rotation angle and the vertical axis represents channels - laser beams. Multiple topologies of relatively shallow CNNs (i.e. 3-5 convolutional layers) are trained and evaluated, using a manually annotated dataset we prepared. The results show significant improvement of performance over the state-of-the-art method by Zhang et al. in terms of speed and also minor improvements in terms of accuracy.

This paper presents a human-carried mapping backpack based on a pair of Velodyne LiDAR scanners. Our system is a universal solution for both large scale outdoor and smaller indoor environments. It benefits from a combination of two LiDAR scanners, which makes the odometry estimation more precise. The scanners are mounted under different angles, thus a larger space around the backpack is scanned. By fusion with GNSS/INS sub-system, the mapping of featureless environments and the georeferencing of resulting point cloud is possible. By deploying SoA methods for registration and the loop closure optimization, it provides sufficient precision for many applications in BIM (Building Information Modeling), inventory check, construction planning, etc. In our indoor experiments, we evaluated our proposed backpack against ZEB-1 solution, using FARO terrestrial scanner as the reference, yielding similar results in terms of precision, while our system provides higher data density, laser intensity readings, and scalability for large environments.

This paper proposes the use of change detection in a multi-view object recognition system in order to improve its flexibility and effectiveness in dynamic environments. Multi-view recognition approaches are essential to overcome problems related to clutter, occlusion or camera noise, but the existing systems usually assume a static environment. The presence of dynamic objects raises another issue - the inconsistencies introduced to the internal scene model. We show that by incorporating the change detection and correction of the inherent scene inconsistencies, we reduce false positive detections by 70% in average for moving objects when tested on the publicly available TUW dataset. To reduce time required for verifying a large set of accumulated object pose hypotheses, we further integrate a clustering approach into the original multi-view object recognition system and show that this reduces computation time by 16%.