Guangcong Zhang

A sparsity-inducing optimization based algorithm for planar patches extraction from noisy point-cloud data

Currently, much of the manual labor needed to generate as-built building information models (BIMs) of existing facilities is spent converting raw point cloud data sets (PCDs) to BIM descriptions. Automating the PCD conversion process can drastically reduce the cost of generating as-built BIMs. Due to the widespread existence of planar structures in civil infrastructures, detecting and extracting planar patches from raw PCDs is a fundamental step in the conversion pipeline from PCDs to BIMs. However, existing methods cannot effectively address both automatically detecting and extracting planar patches from infrastructure PCDs. The existing methods cannot resolve the problem due to the large scale and model complexity of civil infrastructure, or due to the requirements of extra constraints or known information. To address the problem, this article presents a novel framework for automatically detecting and extracting planar patches from large-scale and noisy raw PCDs. The proposed method automatically detects planar structures, estimates the parametric plane models, and determines the boundaries of the planar patches. The first step recovers existing linear dependence relationships amongst points in the PCD by solving a group-sparsity inducing optimization problem. Next, a spectral clustering procedure based on the recovered linear dependence relationships segments the PCD. Then, for each segmented group, model parameters of the extracted planes are estimated via singular value decomposition (SVD) and maximum likelihood estimation sample consensus (MLESAC). Finally, the α-shape algorithm detects the boundaries of planar structures based on a projection of the data to the planar model. The proposed approach is evaluated comprehensively by experiments on two types of PCDs from real-world infrastructures, one captured directly by laser scanners and the other reconstructed from video using structure-from-motion techniques. To evaluate the performance comprehensively, five evaluation metrics are proposed which measure different aspects of performance. Experimental results reveal that the proposed method outperforms the existing methods, in the sense that the method automatically and accurately extracts planar patches from large-scaled raw PCDs without any extra constraints nor user assistance.

Good features to track for visual SLAM

Not all measured features in SLAM/SfM contribute to accurate localization during the estimation process, thus it is sensible to utilize only those that do. This paper describes a method for selecting a subset of features that are of high utility for localization in the SLAM/SfM estimation process. It is derived by examining the observability of SLAM and, being complimentary to the estimation process, it easily integrates into existing SLAM systems. The measure of estimation utility is formulated with temporal and instantaneous observability indices. Efficient computation strategies for the observability indices are described based on incremental singular value decomposition (SVD) and greedy selection for the temporal and instantaneous observability indices, respectively. The greedy selection is near-optimal since the observability index is (approximately) submodular. The proposed method improves localization and data association. Controlled synthetic experiments with ground truth demonstrate the improved localization accuracy, and real-time SLAM experiments demonstrate the improved data association.

Robust Feature Detection, Acquisition and Tracking for Relative Navigation in Space with a Known Target

Copyright

Detecting, Fitting, and Classifying Surface Primitives for Infrastructure Point Cloud Data

This paper presents a novel algorithm for detecting, fitting and classifying the embedded surface primitives from a point cloud dataset (PCD). Given a noisy infrastructure PCD the final output of the algorithm consists of segmented surfaces, their estimated quadric models and corresponding surface classification. Initially, the PCD is down-sampled with a k-d tree structure then segmented via subspace learning. After pose recovery for each segmented group via singular value decomposition, a full quadric model is fit in MLESAC using the direct linear transform for parameter estimation. From the model parameters, the surface is classified from the rank, determinant, and eigenvalues of the parameter matrices. Finally model merging is performed to simplify the results. A real-world PCD of a bridge is used to test the algorithm. The experimental validation of the algorithm demonstrates that the surface primitives are accurately estimated and classified. INTRODUCTION The conditions of a facility during or after construction are not always truthfully represented by the as-designed documentation. In contrast, the as-built building information models (BIMs) do describe an existing facility more accurately. A typical process to generate as-built BIMs uses raw PCD from remote-sensing or photogrammetry as input and outputs an information-rich object oriented models. However, currently generating as-built BIMs is cost-prohibitive, because the conversion from the raw PCDs to the geometric models is a very time-consuming manual process. Therefore, automating this conversion process is of great significance for greater use of as-built BIMs. To address this problem, Zhang (Zhang, G., et al, 2012) proposed a sparsityinducing optimization based algorithm to extract planar patches from noisy PCDs. The algorithm first segments the PCD into linear subspaces and then performs robust model estimation with planar models. This approach has been demonstrated to be effective for real-world infrastructure PCDs. However, it only works for planar surfaces. In contrast, this work seeks to extract more kinds of structures from infrastructure PCDs. This work focuses on processing PCDs instead of the registered images to make the algorithm applicable to not only image-based reconstructed PCDs but also PCDs from laser scanners. The algorithm is designed to detect, fit, and

Cooperative Relative Navigation for Space Rendezvous and Proximity Operations using Controlled Active Vision

This work aims to solve the problem of relative navigation for space rendezvous and proximity operations using a monocular camera in a numerically efficient manner. It is assumed that the target spacecraft has a special pattern to aid the task of relative pose estimation, and that the chaser spacecraft uses a monocular camera as the primary visual sensor. In this sense, the problem falls under the category of cooperative relative navigation in orbit. While existing systems for cooperative localization with fiducial markers allow full six-degrees-of-freedom pose estimation, the majority of them are not suitable for in-space cooperative navigation especially when involving a small-size chaser spacecraft, due to their computational cost. Moreover, most existing fiducial-based localization methods are designed for ground-based applications with limited range e.g., ground robotics, augmented reality, and their performance deteriorates under large-scale changes, such as those encountered in space applications. Using an adaptive visual algorithm, we propose an accurate and numerically efficient approach for real-time vision-based relative navigation, especially designed for space robotics applications. The proposed method achieves low computational cost and high accuracy and robustness via the following innovations: first, an adaptive visual pattern detection scheme based on the estimated relative pose is proposed, which improves both the efficiency of detection and the accuracy of pose estimates; second, a parametric blob detector called Box-LoG is used, which is computationally efficient; and third, a fast and robust algorithm is introduced, which jointly solves the data association and pose estimation problems. In addition to having an accuracy comparable to state-of-the-art cooperative localization algorithms, our method demonstrates a significant improvement in speed and robustness for scenarios with large range changes. A vision-based closed-loop experiment using the Autonomous Spacecraft Testing of Robotic Operations in Space ASTROS testbed demonstrates the performance benefits of the proposed approach.

Optimally observable and minimal cardinality monocular SLAM

This paper utilizes system observability to guide monocular SLAM. Instead of providing all measured features then performing data-driven outlier rejection (such as with RANSAC), we propose to identify only the minimal subset of features which form an optimally observable SLAM subsystem for localization. Modeling the SLAM system as a discrete time system with piece-wise linear SE〈3〉 motion, complete observability conditions are derived and a means to test the observability conditioning of candidate feature point groupings is proposed. Based on the conditioning, an efficient algorithm for picking the optimally observable feature subset is derived by incorporating the image geometric measures. The proposed monocular SLAM algorithm, called Optimally Observable and Minimal Cardinality (OOMC) SLAM is formulated as an EKF process. OOMC SLAM is first validated using a 6-DOF localization experiment; the results demonstrate accuracy comparable to the state-of-art SLAM algorithm with significantly improved computational efficiency. A longer sequence on a 620-meter trajectory is also tested. The algorithm achieves 0.9178% relative error against the GPS ground truth.

Automatic Generation of As-Built Geometric Civil Infrastructure Models from Point Cloud Data

Converting remote sensing point cloud data (PCD) into solid CAD models consisting of civil infrastructure components is a crucial step in generating the asbuilt building information models. Previous research has enabled automatic generation of surface primitives from raw PCDs. However, the fully automatic conversion from surface primitives to infrastructure component models remains an unsolved problem. In this work, an automatic and linear-runtime approach is presented which generates the as-built infrastructure component models by recognizing the solid CAD entities and learning the infrastructure component labels from the fitted surface primitives. The algorithm utilizes a decision tree with the following decision variables: the type, parametric model, orientation, and mutual geometric relations of the fitted surface primitives. The decision tree is trained with easily generated synthetic data and is applied to query real-world data with complexity O(1). The output of the solid entities includes cuboid, cylinder, ball, etc. and the infrastructure component labels such as columns, caps, deck, beams, etc. The algorithm is tested with various PCDs modeling real bridges.

Learning binary features online from motion dynamics for incremental loop-closure detection and place recognition

This paper proposes a simple yet effective approach to learn visual features online for improving loop-closure detection and place recognition, based on bag-of-words frameworks. The approach learns a codeword in the bag-of-words model from a pair of matched features from two consecutive frames, such that the codeword has temporally-derived perspective invariance to camera motion. The learning algorithm is efficient: the binary descriptor is generated from the mean image patch, and the mask is learned based on discriminative projection by minimizing the intra-class distances among the learned feature and the two original features. A codeword is generated by packaging the learned descriptor and mask, with a masked Hamming distance defined to measure the distance between two codewords. The geometric properties of the learned codewords are then mathematically justified. In addition, hypothesis constraints are imposed through temporal consistency in matched codewords, which improves precision. The approach, integrated in an incremental bag-of-words system, is validated on multiple benchmark data sets and compared to state-of-the-art methods. Experiments demonstrate improved precision/recall outperforming state of the art with little loss in runtime.

A Sparsity-Inducing Optimization Algorithm for the Extraction of Planar Structures in Noisy Point-Cloud Data

Most of the manual labor needed to create the geometric building information model (BIM) of an existing facility is spent converting raw point cloud data (PCD) to a BIM description. Automating this process would drastically reduce the modeling cost. Surface extraction from PCD is a fundamental step in this process. Compact modeling of redundant points in PCD as a set of planes leads to smaller file size and fast interactive visualization on cheap hardware. Traditional approaches for smooth surface reconstruction do not explicitly model the sparse scene structure or significantly exploit the redundancy. This paper proposes a method based on sparsity-inducing optimization to address the planar surface extraction problem. Through sparse optimization, points in PCD are segmented according to their embedded linear subspaces. Within each segmented part, plane models can be estimated. Experimental results on a typical noisy PCD demonstrate the effectiveness of the algorithm.

Efficient Closed-Loop Detection and Pose Estimation for Vision-Only Relative Localization in Space with a Cooperative Target

Copyright