The capability to navigate accurately is one of the features, that a mobile robot should have to be able to perform tasks autonomously. In a GPS/GNSS-denied environment, for example inside buildings, localization of a mobile platform is an especially challenging problem. In such cases, to provide a robot with the ability to determine its position and to analyze its surroundings, Simultaneous Localization and Mapping (SLAM) algorithms could be implemented. In the article, we present a SLAM system that uses a Kalman filter together with data gathered by a 2D LiDAR. Our approach applies the ICP algorithm to calculate the localization and employs clustering and shape recognition technics to build the map of the environment. The article contains a detailed description of the individual elements of the proposed SLAM solution. Furthermore, it presents the results of experiments during which the system was validated.
The article contains an analysis of potential prospects of simultaneous localization and mapping (SLAM) algorithms application in imagery intelligence (IMINT). The first part of the paper presents a detailed description of the SLAM problem. Diverse solutions to the simultaneous localization and mapping problem and related research over the years are presented. The most promising of SLAM approaches are pointed out. To facilitate SLAM analysis, the problem is partitioned into three parts. First, various SLAM estimation techniques are characterized. A mathematical theory behind the usage of parametric filters, non-parametric filters, and least squares method is presented. Further, differences between SLAM algorithms are described in terms of various sensors used on-board SLAM platforms for the examination of the environment. The examination is commonly addressed as landmark extraction. A separate part of the paper discusses the image processing in SLAM. The last part of the SLAM analysis is dedicated to various approaches to map presentation. Further, the properties of SLAM techniques are characterized in terms of their potential benefits to IMINT. Prospects of increased efficiency, accuracy and safety of intelligence gathering process are discussed.