External calibration of a camera to a laser rangefinder is a common pre-requisite on today’s multi-sensor mobile robot platforms. However, the process of doing so is relatively poorly documented and almost always time-consuming. This document outlines an easy and portable technique for external calibration of a camera to a laser rangefinder. It describes the usage of the Laser-Camera Calibration Toolbox (LCCT), a Matlab R ○-based graphical user interface that is meant to accompany this document and facilitates the calibration procedure. We also summarize the math behind its development. The software is accessible online atwww.cs.cmu.edu/˜ranjith/lcct.html, as well as at the VMR Lab Software page atwww.cs.cmu.edu/˜vmr/software/ software.html.

In this paper, we describe a new approach for the extrinsic calibration of a camera with a 3D laser range finder, that can be done on the fly. This approach does not require any calibration object. Only few point correspondences (at least 4) are used, which are manually selected by the user from a scene viewed by the two sensors. The proposed method relies on a novel technique to visualize the range information obtained from a 3D laser scanner. This technique converts the visually ambiguous 3D range information into a 2D map where natural features of a scene are highlighted. These features represent depth discontinuities and direction changes in the range image. We show that by enhancing the features the user can easily find the corresponding points of the camera image points. Therefore, visually identifying laser-camera correspondences becomes as easy as image pairing. Once point correspondences are given, extrinsic calibration is done using the well-known PnP algorithm followed by a non-linear refinement process. We show the performance of our approach through experimental results. In these experiments, we will use an omnidirectional camera. The implication of this method is important because it brings 3D computer vision systems out of the laboratory and into practical use.

Abstract — This paper presents a novel human detection method based on a Bayesian fusion approach using laser range data and camera images. Laser range data analysis groups data points with a novel graph cutting method. Therefore, it computes a belief to each cluster based on the evaluation of multidimensional features that describe geometrical properties. A person detection algorithm based on dense overlapping grid of Histograms of Oriented Gradients (HOG) is processed on the image area determined by each laser cluster. The selection of HOG features and laser features is obtained through a learning process based on a cascade of linear Support Vector Machines (SVM). A technique to obtain conditional probabilities from a cascade of SVMs is here proposed in order to combine the two information together. The resulting human detection consists in a rich information that takes into account the distance of the cluster and the confidence level of both detection methods. We demonstrate the performance of this work on real-world data and different environments.

www.elsevier.com/locate/robot This paper presents Scan-SLAM, a new generalization of simultaneous localization and mapping (SLAM). SLAM implementations based on extended Kalman filter (EKF) data fusion have traditionally relied on simple geometric models for defining landmarks. This limits EKF-SLAM to environments suited to such models and tends to discard much potentially useful data. The approach presented in this paper is a marriage of EKF-SLAM and scan correlation. Landmarks are no longer defined by analytical models; instead they are defined by templates composed of raw sensed data. These templates can be augmented as more data become available so that the landmark definition improves with time. A new generic observation model is derived that is generated by scan correlation, and this permits stochastic location estimation for landmarks with arbitrary shape within the Kalman filter framework. The statistical advantages of an EKF representation are augmented with the general applicability of scan matching. Scan matching also serves to enhance data association reliability by providing a shape metric for landmark disambiguation. Experimental results in an outdoor environment are presented which validate the algorithm. c ○ 2006 Elsevier B.V. All rights reserved. Keywords: Simultaneous localisation and mapping (SLAM); EKF-SLAM; Scan correlation

Abstract. This paper presents a general probabilistic framework for multisensor multi-class object recognition based on Conditional Random Fields (CRFs) trained with virtual evidence boosting. The learnt representation models spatial and temporal relationships and is able to integrate arbitrary sensor information by automatically extracting features from data. We demonstrate the benefits of modelling spatial and temporal relationships for the problem of detecting seven classes of objects using laser and vision data in outdoor environments. Additionally, we show how this framework can be used with partially labeled data, thereby significantly reducing the burden of manual data annotation. 1

Abstract — Secure driving in dynamic environments is an application requiring a number of premises. First of all it needs a perception system able to detect and register obstacles in the vicinity of the robot. Those obstacles are mapped and passed to a motion planner able to calculate a path considering the global objective as well as locally collision free trajectories. Finally, as the calculated path is only guaranteed to be collision free within certain boundaries, it needs a precise path following module commanding the vehicle to follow the calculated path precisely. In this paper we will show how we tackle those three primary requirements for safe driving in dynamic environments: On the perception side we use three main sensors to perceive environment information. For the mapping of arbitrary obstacles we use a setup of three different kinds of sensors. One IBEO Alasca XT Laser Scanner mounted at the front of the vehicle to provide short and long range object data, and two Sick LMS 291 looking down from the upper corners of the car securing cornering. Form those data a local traversability map is calculated that is passed to the motion planner. Another software module uses a sensor fusion approach to detect pedestrians: a laser scans analysis is computed to create weighted regions of interest in the scene; within those regions a vision algorithm based on an advanced cascade of classifiers of fast image features is applied to precisely detect people in the perceived environment. The navigation side is using a combination of a global Field D-Star planner combined with a local path planner that forwardsimulates trajectories and checks those for collisions. Finally the desired vehicle trajectory is executed by the path following algorithm using a sliding controller to keep the car on the secure track. The paper concludes with experimental results from autonomous driving in different scenarios. I.

Abstract — This paper presents a method for estimating the six-degrees-of-freedom transformation between a camera and the body of the robot on which it is rigidly attached. The robot maneuvers in front of a planar mirror, allowing the camera to observe fiducial features on the robot from several vantage points. Exploiting these measurements, we form a maximum-likelihood estimate of the camera-to-body transformation, without assuming prior knowledge of the robot motion or of the mirror configuration. Additionally, we estimate the mirror configuration with respect to the camera for each image. We validate the accuracy and correctness of our method with simulations and real-world experiments. I.

Abstract — We propose a new method for extrinsic calibration of a line-scan LIDAR with a perspective projection camera. Our method is a closed-form, minimal solution to the problem. The solution is a symbolic template found via variable elimination and the multi-polynomial Macaulay resultant. It does not require initialization, and can be used in an automatic calibration setting when paired with RANSAC and least-squares refinement. We show the efficacy of our approach through a set of simulations and a real calibration. I.

Abstract — In this paper, we examine the problem of extrinsic calibration of multiple LIDARs on a mobile vehicle platform. To achieve fully automated and on-line calibration, the original non-linear calibration model is reformulated as a second-order cone program (SOCP). This provides an advantage over more standard linearized approaches in that a priori information such as a default LIDAR calibration, calibration tolerances, etc., can be readily modeled. Furthermore, in contrast to general non-linear methods, the SOCP relaxation is convex, returns a global minimum, and can be solved very quickly using modern interior point methods (IPM). This enables the calibration to be estimated on-line for multiple LIDARs simultaneously. Experimental results are provided where the approach is used to successfully calibrate a pair of Sick LMS291-S14 LIDARs mounted on a mobile vehicle platform. These showed the SOCP

We present a novel method for fully automated exterior calibration of a 2D scanning laser range sensor that attains accurate pose with respect to a fixed 3D reference frame. This task is crucial for applications that attempt to recover self-consistent 3D environment maps and produce accurately registered or fused sensor data. A key contribution of our approach lies in the design of a class of calibration target objects whose pose can be reliably recognized from a single observation (i.e. from one 2D range data stripe). Unlike other techniques, we do not require simultaneous camera views or motion of the sensor, making our approach simple, flexible and environment-independent. In this paper we illustrate the target geometry and derive the relationship between a single 2D range scan and the 3D sensor pose. We describe an algorithm for closed-form solution of the 6 DOF pose that minimizes an algebraic error metric, and an iterative refinement scheme that subsequently minimizes geometric error. Finally, we report performance and stability of our technique on synthetic and real data sets, and demonstrate accuracy within 1 degree of orientation and 3 cm of position in a realistic configuration. 1