Autonomous vehicles are expected to play a key role in the future of urban transportation systems, as they offer potential for additional safety, increased productivity, greater accessibility, better road efﬁciency, and positive impact on the environment. Research in autonomous systems has seen dramatic advances in recent years, due to the increases in available computing power and reduced cost in sensing and computing technologies, resulting in maturing technological readiness level of fully autonomous vehicles. The objective of this paper is to provide a general overview of the recent developments in the realm of autonomous vehicle software systems. Fundamental components of autonomous vehicle software are reviewed, and recent developments in each area are discussed.

Perception, Planning, Control, and Coordination for Autonomous Vehicles

This paper introduces a novel, region-growing algorithm for the fast surface patch segmentation of three-dimensional point clouds of urban environments. The proposed algorithm is composed of two stages based on a coarse-to-fine concept. First, a region-growing step is performed on an octree-based voxelized representation of the input point cloud to extract major (coarse) segments. The output is then passed through a refinement process. As part of this, there are two competing factors related to voxel size selection. To balance the constraints, an adaptive octree is created in two stages. Empirical studies on real terrestrial and airborne laser scanning data for complex buildings and an urban setting show the proposed approach to be at least an order of magnitude faster when compared to a conventional region growing method and able to incorporate semantic-based feature criteria, while achieving precision, recall, and fitness scores of at least 75% and as much as 95%.

Octree-based region growing for point cloud segmentation

Object-based point cloud analysis (OBPA) is useful for information extraction from airborne LiDAR point clouds. An object-based classification method is proposed for classifying the airborne LiDAR point clouds in urban areas herein. In the process of classification, the surface growing algorithm is employed to make clustering of the point clouds without outliers, thirteen features of the geometry, radiometry, topology and echo characteristics are calculated, a support vector machine (SVM) is utilized to classify the segments, and connected component analysis for 3D point clouds is proposed to optimize the original classification results. Three datasets with different point densities and complexities are employed to test our method. Experiments suggest that the proposed method is capable of making a classification of the urban point clouds with the overall classification accuracy larger than 92.34% and the Kappa coefficient larger than 0.8638, and the classification accuracy is promoted with the increasing of the point density, which is meaningful for various types of applications.

SVM-Based Classification of Segmented Airborne LiDAR Point Clouds in Urban Areas

As an important component of urban vegetation, street trees play an important role in maintenance of environmental quality, aesthetic beauty of urban landscape, and social service for inhabitants. Acquiring accurate and up-to-date inventory information for street trees is required for urban horticultural planning, and municipal urban forest management. This paper presents a new Voxel-based Marked Neighborhood Searching (VMNS) method for efficiently identifying street trees and deriving their morphological parameters from Mobile Laser Scanning (MLS) point cloud data. The VMNS method consists of six technical components: voxelization, calculating values of voxels, searching and marking neighborhoods, extracting potential trees, deriving morphological parameters, and eliminating pole-like objects other than trees. The method is validated and evaluated through two case studies. The evaluation results show that the completeness and correctness of our method for street tree detection are over 98%. The derived morphological parameters, including tree height, crown diameter, diameter at breast height (DBH), and crown base height (CBH), are in a good agreement with the field measurements. Our method provides an effective tool for extracting various morphological parameters for individual street trees from MLS point cloud data.

A Voxel-Based Method for Automated Identification and Morphological Parameters Estimation of Individual Street Trees from Mobile Laser Scanning Data

Plane segmentation is a basic task in the automatic reconstruction of indoor and urban environments from unorganized point clouds acquired by laser scanners. As one of the most common plane-segmentation methods, standard Random Sample Consensus (RANSAC) is often used to continually detect planes one after another. However, it suffers from the spurious-plane problem when noise and outliers exist due to the uncertainty of randomly sampling the minimum subset with 3 points. An improved RANSAC method based on Normal Distribution Transformation (NDT) cells is proposed in this study to avoid spurious planes for 3D point-cloud plane segmentation. A planar NDT cell is selected as a minimal sample in each iteration to ensure the correctness of sampling on the same plane surface. The 3D NDT represents the point cloud with a set of NDT cells and models the observed points with a normal distribution within each cell. The geometric appearances of NDT cells are used to classify the NDT cells into planar and non-planar cells. The proposed method is verified on three indoor scenes. The experimental results show that the correctness exceeds 88.5% and the completeness exceeds 85.0%, which indicates that the proposed method identifies more reliable and accurate planes than standard RANSAC. It also executes faster. These results validate the suitability of the method.

An Improved RANSAC for 3D Point Cloud Plane Segmentation Based on Normal Distribution Transformation Cells

Lidar (light detection and ranging) data implicitly contains abundant three-dimensional spatial information. The segmentation of lidar point clouds is the key procedure for transforming implicit spatial information into explicit spatial information. Common criteria used for point cloud segmentation are proximity and coherence of point distribution. An effective segmentation algorithm may apply various steps or combinations of criteria depending on the application. This paper proposes a four-step segmentation method for lidar point clouds to deliver incremental segmentation results. Segmentation results of each step can provide the fundamental data for the next step. In the first step, the input point cloud is organised into an octree-structured voxel space, in which the point neighbourhood is established. In the second step, connected voxels which are not empty are grouped to obtain grouped points based on proximity. The third step is a coplanar point segmentation based on both coherence and proximity, which was performed on each point group obtained in the second step. Finally, neighbouring coplanar point groups are merged into “co-surface” point groups based on the criteria of plane connection and intersection. This scheme enables an incremental retrieval and analysis of a large lidar data-set. Experimental results demonstrate the effectiveness of the segmentation algorithm in handling both airborne and terrestrial lidar data. It is anticipated that the incremental segmentation results will be useful for object modelling using lidar data.

Resume

Les donnees lidar contiennent implicitement une abondante information spatiale tridimensionnelle. La segmentation de nuages de points lidar est la procedure qui permet de transformer cette information spatiale implicite en une information spatiale explicite. Les criteres couramment utilises pour la segmentation de nuages de points lidar sont la proximite et la coherence de la distribution de points. Un algorithme efficace de segmentation peut comprendre plusieurs etapes correspondant a des combinaisons de criteres qui dependent de l’application visee. Cet article propose une methode de segmentation en quatre etapes, concue pour fournir a partir de nuages de points des resultats incrementiels de segmentation. Les resultats de segmentation obtenus a chaque etape fournissent les donnees de base pour l’etape suivante. A la premiere etape, le nuage de points initial est reorganise dans un espace de voxels structure en arbres octaires. A la deuxieme etape, les voxels connexes qui ne sont pas vides sont relies pour regrouper les points correspondants sur un critere de proximite. La troisieme etape est une segmentation de points coplanaires basee a la fois sur des criteres de coherence et de proximite, et appliquee a chacun de groupes de points obtenus a la deuxieme etape. Finalement, les groupes coplanaires adjacents sont regroupes en groupes de points cosurfaciques sur des criteres de connexion et intersection de plans. Cette procedure permet le traitement et l’analyse incrementiels de donnees lidar volumineuses. Les resultats experimentaux montrent l’efficacite de l’algorithme de segmentation, pour des donnees lidar aeroportees aussi bien que terrestres. On peut s’attendre a ce que les resultats de la segmentation incrementielle permettent la modelisation d’objets a partir de donnees lidar.

Zusammenfassung

Lidardaten beinhalten implizit reichlich drei-dimensionale raumliche Information. Eine Segmentierung von Lidarpunktwolken ist ein zentrales Element in der Transformation von impliziter in explizite raumliche Information. Ublicherweise werden die Kriterien Nachbarschaft und Koharenz bei der Segmentierung von Punktwolken eingesetzt. Ein effektiver Segmentierungsalgorithmus kann verschiedene Schritte oder Kombinationen von Kriterien in Abhangigkeit von der Anwendung einsetzen. In diesem Artikel wird eine Segmentierungsmethode fur Lidarpunktwolken, basierend auf vier Schritten vorgeschlagen, um inkrementelle Segmentierungsergebnisse zu erhalten. In einem ersten Schritte wird die Eingangspunktwolke in einen Voxel-Raum mit Octree-Struktur organisiert, wobei die Punktnachbarschaft eingefuhrt wird. In einem zweiten Schritt werden verbundene Voxel, die nicht leer sind, gruppiert, um gruppierte Punkte nach dem Nachbarschaftskriterium zu erhalten. Der dritte Schritt ist eine koplanare Punktsegmentierung, die sowohl auf Koharenz, als auch auf Nachbarschaft basiert, und die auf jede Punktgruppe aus Schritt zwei angewandt wird. Zuletzt werden koplanare Gruppen verbunden, die auf einer gemeinsamen Oberflache liegen, basierend auf Kriterien zur Ebenenverbindung und -schnitt. Dieses beschriebene Schema erlaubt eine inkrementelle Ableitung und Analyse von grosen Lidardatensatzen. Experimentelle Untersuchungen demonstrieren die Effektivitat des Segmentierungsalgorithmus fur Lidardaten aus flugzeuggestutztem oder terrestrischem Laserscanning. Aller Voraussicht nach konnen die Ergebnisse der inkrementellen Segmentierung fur die Objektmodellierung mit Lidardaten nutzlich sein.

Resumen

En los datos lidar (Light Detection and Ranging) hay implicita una abundante informacion espacial tridimensional. De modo que la segmentacion de nubes de puntos es un procedimiento fundamental para transformar dicha informacion espacial implicita en explicita. Los criterios mas utilizados para la segmentacion de las nubes de puntos son la proximidad y la coherencia de la distribucion de puntos. Un algoritmo efectivo de segmentacion puede constar de varios pasos o de aplicar combinaciones de criterios, dependiendo de la aplicacion. Este articulo propone un metodo de segmentacion en cuatro pasos que proporciona resultados graduales a lo largo de la segmentacion y en donde los resultados obtenidos en cada paso son los datos base para el siguiente. En el primer paso, la nube de puntos se organiza en un espacio voxel estructurado mediante un arbol octal en donde se define la vecindad del punto. En el segundo paso, los voxel conectados que no esten vacios se agrupan para obtener aglomerados de puntos en funcion de la proximidad. El tercer paso consiste en realizar una segmentacion coplanaria de puntos basada en la coherencia y la proximidad, de cada uno de los grupos de puntos identificado en el segundo paso. Finalmente, los grupos coplanarios vecinos se unen en grupos de puntos sobre una misma superficie basandose en criterios de conexion e interseccion de planos. Este procedimiento permite la extraccion y analisis gradual de un gran conjunto de datos lidar. Los resultados experimentales demuestran la efectividad del algoritmo para datos lidar aerotransportado y terrestre, lo que permite anticipar que los resultados de la segmentacion gradual podran utilizarse para el modelado de objetos con datos lidar.

Incremental segmentation of lidar point clouds with an octree-structured voxel space

Lidar (light detection and ranging) point cloud data contain abundant three-dimensional (3D) information. Dense distribution of scanned points on object surfaces prominently implies surface features. Particularly, plane features commonly appear in a typical lidar dataset of artificial structures. To explore implicitly contained spatial information, this study developed an automatic scheme to segment a lidar point cloud dataset into coplanar point clusters. The central mechanism of the proposed method is a split-and-merge segmentation based on an octree structure. Plane fitting serves as an engine in the mechanism that evaluates how well a group of points fits to a plane. Segmented coplanar points and derived parameters of their best-fit plane are obtained through the process. This paper also provides algorithms to derive various geometric properties of segmented coplanar points, including inherent properties of a plane, intersections of planes, and properties of point distribution on a plane. Several successful cases of handling airborne and terrestrial lidar data as well as a combination of the two are demonstrated. This method should improve the efficiency of object modelling using lidar data.

Miao Wang

Papers

Incremental segmentation of lidar point clouds with an octree-structured voxel space

Automatic Segmentation of Lidar Data into Coplanar Point Clusters Using an Octree-Based Split-and-Merge Algorithm