A fast pose-graph relaxation technique is presented for enhancing the consistency of three-dimensional (3D) maps created by registering large planar surface patches. The surface patches are extracted from point clouds sampled from a 3D range sensor. The plane-based registration method offers an alternative to the state-of-theart algorithms and provides advantages in terms of robustness, speed, and storage. One of its features is that it results in an accurate determination of rotation, although a lack of predominant surfaces in certain directions may result in translational uncertainty in those directions. Hence, a loop-closing and relaxation problem is formulated that gains significant speed by relaxing only the translational errors and utilizes the full-translation covariance determined during pairwise registration. This leads to a fast 3D simultaneous localization and mapping suited for online operations. The approach is tested in two disaster scenarios that were mapped at the NIST 2008 Response Robot Evaluation Exercise in Disaster City, Texas. The two data sets from a collapsed car park and a flooding disaster consist of 26 and 70 3D scans, respectively. The results of these experiments show that our approach can generate 3D maps without motion estimates by odometry and that it outperforms iterative closest point-based mapping with respect to speed and robustness. C

Simultaneous Localization and Mapping (SLAM) has focused on noisy but unique data associations resulting in linear Gaussian uncertainty models. However, a unique decision is often not possible using only local information, giving rise to ambiguities that have to be resolved globally during optimization. To solve this problem, the pose graph data structure is extended here by multimodal constraints modeled by mixtures of Gaussians (MoG). Furthermore, optimization methods for this novel formulation are introduced, namely (a) robust iteratively reweighted least squares, and (b) Prefilter Stochastic Gradient Descent (SGD) where a preprocessing step determines globally consistent modes before applying SGD. In addition, a variant of the Prefilter method (b) is introduced in form of (c) Prefilter Levenberg-Marquardt. The methods are compared with traditional state-of-the-art optimization methods including (d) Stochastic Gradient Descent and (e) Levenberg-Marquardt as well as (f) Particle filter SLAM and with (g) an optimal exhaustive algorithm. Experiments show that ambiguities significantly impact state-of-the-art methods, and that the novel Prefilter methods (b) and (c) perform best. This is further substantiated with experiments using real-world data. To this end, a method to generate MoG constraints from a plane-based registration algorithm is introduced and used for 3D SLAM under ambiguities.

No abstract

This article addresses fast 3D mapping by a mobile robot in a predominantly planar environment. It is based on a novel pose registration algorithm based entirely on matching features composed of plane-segments extracted from point-clouds sampled from a 3D sensor. The approach has advantages in terms of robustness, speed and storage as compared to the voxel based approaches. Unlike previous approaches, the uncertainty in plane parameters is utilized to compute the uncertainty in the pose computed by scan-registration. The algorithm is illustrated by creating a full 3D model of a multi-level robot testing arena.

No abstract