Robotica: page 1 of 19. © Cambridge University Press 2013 doi:10.1017/S0263574713001070 Topological simultaneous localization and 1 mapping: a survey 2 Jaime Boal ∗ , ´ Alvaro S´ anchez-Miralles and 3 ´ Alvaro Arranz 4 Institute for Research in Technology (IIT),ICAI School of Engineering, Comillas Pontifical 5 University, Madrid, Spain 6 (Accepted October 29, 2013) 7 SUMMARY 8 One of the main challenges in robotics is navigating autonomously through large, unknown, and 9 unstructured environments. Simultaneous localization and mapping (SLAM) is currently regarded 10 as a viable solution for this problem. As the traditional metric approach to SLAM is experiencing 11 computational difficulties when exploring large areas, increasing attention is being paid to topological 12 SLAM, which is bound to provide sufficiently accurate location estimates, while being significantly 13 less computationally demanding. This paper intends to provide an introductory overview of the most 14 prominent techniques that have been applied to topological SLAM in terms of feature detection, map 15 matching, and map fusion. 16 17 KEYWORDS: Mobile robots; SLAM; Topological modeling of robots; Feature detection; Robot 18 localization 19 1. Introduction 20 Mobile robotics’ ultimate aim is to develop fully autonomous entities capable of performing rather 21 complicated tasks, without the need for human intervention, during extended periods of time. Over 22 the past three decades, this objective has constantly faced harsh difficulties, which have hindered 23 progress. The most recurrent issues in the literature, which are yet to be completely resolved, are 24 stated below. 25 A mobile robot must be able to navigate through the environment in order to achieve its goals. 26 According to Leonard and Durrant-Whyte, 63 this general problem can be summarized in three 27 questions: “Where am I?,” “Where am I going?,” and “How should I get there?” The first question 28 addresses the localization problem, which intends to estimate the robot’s pose (i.e., location and 29 orientation) using data gathered by distinct sensors and knowledge of previous locations. However, 30 the presence of noisy sensor measurements makes this problem harder than it may seem at first 31 sight. The precision with which this problem is solved decisively affects the answer to the other two 32 questions, as it is necessary to localize oneself in the environment to safely interact with it, decide 33 what the following step should be, and how to accomplish it. 34 During the localization process, a robot must resort to some kind of reference system; in other 35 words, it requires a map. The extensive research survey carried out by Thrun 110 collects the main 36 open issues concerning robotic mapping, which are succinctly presented henceforth. Currently, there 37 are robust methods for mapping structured, static, and bounded environments, whereas mapping 38 unstructured, dynamic, or large-scale unknown environments remains largely an unsolved problem. 39 According to Thrun, 110 the robotic mapping problem is “that of acquiring a spatial model of a 40 robot’s environment.” To this end, robots must be equipped with sensors that enable them to perceive 41 the outside world. Once again, sensor errors and range limitations pose a great difficulty. 42 The first challenge in robotic mapping develops from the measurement noise. Usually, this issue 43 can be overcome if the noise is statistically independent, as it can be canceled out performing enough 44 measurements. Unfortunately, this does not always occur in robotic mapping because, whenever 45 * Corresponding author. E-mail: jaime.boal@iit.upcomillas.es