Abstract-Simultaneous localization and mapping (SLAM) consists in the concurrent construction of a model of the environment (the map), and the estimation of the state of the robot moving within it. The SLAM community has made astonishing progress over the last 30 years, enabling large-scale real-world applications and witnessing a steady transition of this technology to industry. We survey the current state of SLAM and consider future directions. We start by presenting what is now the de-facto standard formulation for SLAM. We then review related work, covering a broad set of topics including robustness and scalability in long-term mapping, metric and semantic representations for mapping, theoretical performance guarantees, active SLAM and exploration, and other new frontiers. This paper simultaneously serves as a position paper and tutorial to those who are users of SLAM. By looking at the published research with a critical eye, we delineate open challenges and new research issues, that still deserve careful scientific investigation. The paper also contains the authors' take on two questions that often animate discussions during robotics conferences: Do robots need SLAM? and Is SLAM solved?

Fig. 1: We demonstrate object-oriented semantic mapping using RGB-D data that scales from small desktop environments (left) to offices (middle) and whole labs (right). The pictures show 3D map structures with objects colored according to their semantic class. We do not merely project semantic labels for individual 3D points, but rather maintain objects as the central entity of the map, freeing it from the requirement for a-priori 3D object models in [1]. To achieve this, our system creates and extends 3D object models while continuously mapping the environment. Object detection and classification is performed using a Convolutional Network, while an unsupervised 3D segmentation algorithm assigns a segment of 3D points to every object detection. These segmented object detections are then either fused with existing objects, or added as a new object to the map. ORB-SLAM2 provides a global SLAM solution that enables us to reconstruct a 3D model of the environment that contains both non-object structure and objects of various types.Abstract-For intelligent robots to interact in meaningful ways with their environment, they must understand both the geometric and semantic properties of the scene surrounding them. The majority of research to date has addressed these mapping challenges separately, focusing on either geometric or semantic mapping. In this paper we address the problem of building environmental maps that include both semantically meaningful, object-level entities and point-or mesh-based geometrical representations. We simultaneously build geometric point cloud models of previously unseen instances of known object classes and create a map that contains these object models as central entities. Our system leverages sparse, feature-based RGB-D SLAM, image-based deep-learning object detection and 3D unsupervised segmentation.

Long term autonomous mobile robot operation requires considering place recognition decisions with great caution. A single incorrect decision that is not detected and reconsidered can corrupt the environment model that the robot is trying to build and maintain. This work describes a consensus-based approach to robust place recognition over time, that takes into account all the available information to detect and remove past incorrect loop closures. The main novelties of our work are: (1) the ability of realizing that, in light of new evidence, an incorrect past loop closing decision has been made; the incorrect information can be removed thus recovering the correct estimation with a novel algorithm; (2) extending our proposal to incremental operation; and (3) handling multi-session, spatially related or unrelated scenarios in a unified manner. We demonstrate our proposal, the RRR algorithm, on different odometry systems, e.g. visual or laser using different front-end loop closing techniques. For our experiments we use the efficient graph optimization framework g2o as back-end. We back our claims with several experiments carried out on real data, in single and multi-session experiments showing better results than those obtained by state of the art methods, comparisons against whom are also presented.

Abstract-Long term autonomy in robots requires the ability to reconsider previously taken decisions when new evidence becomes available. Loop closing links generated by a place recognition system may become inconsistent as additional evidence arrives. This paper is concerned with the detection and exclusion of such contradictory information from the map being built, in order to recover the correct map estimate. We propose a novel consistency based method to extract the loop closure regions that agree both among themselves and with the robot trajectory over time. We also assume that the contradictory loop closures are inconsistent among themselves and with the robot trajectory. We support our proposal, the RRR algorithm, on well-known odometry systems, e.g. visual or laser, using the very efficient graph optimization framework g2o as back-end. We back our claims with several experiments carried out on real data.