Abstract
Simultaneous Localization and Mapping is now widely adopted by many applications, and researchers have produced very dense literature on this topic. With the advent of smart devices, embedding cameras, inertial measurement units, visual SLAM (vSLAM), and visual-inertial SLAM (viSLAM) are enabling novel general public applications. In this context, this paper conducts a review of popular SLAM approaches with a focus on vSLAM/viSLAM, both at fundamental and experimental levels. It starts with a structured overview of existing vSLAM and viSLAM designs and continues with a new classification of a dozen main state-of-the-art methods. A chronological survey of viSLAM’s development highlights the historical milestones and presents more recent methods into a classification. Finally, the performance of vSLAM is experimentally assessed for the use case of pedestrian pose estimation with a handheld device in urban environments. The performance of five open-source methods Vins-Mono, ROVIO, ORB-SLAM2, DSO, and LSD-SLAM is compared using the EuRoC MAV dataset and a new visual-inertial dataset corresponding to urban pedestrian navigation. A detailed analysis of the computation results identifies the strengths and weaknesses for each method. Globally, ORB-SLAM2 appears to be the most promising algorithm to address the challenges of urban pedestrian navigation, tested with two datasets.
Highlights
The Simultaneous Localization and Mapping (SLAM) problem has been one of the most active research subjects since its formulation in the 1980s [1, 2]
We conducted a review of important SLAM approaches and detailed the core notions of Visual SLAM (vSLAM) and visual-inertial SLAM (viSLAM) along with the different existing designs
We linked this theoretical survey with a historical overview to identify the main milestones in SLAM evolution divided into three main periods
Summary
The Simultaneous Localization and Mapping (SLAM) problem has been one of the most active research subjects since its formulation in the 1980s [1, 2]. SLAM’s goal is to obtain a global and consistent estimate of a device’s path while reconstructing a map of the surrounding environment The coupling between these two tasks, initially considered as the core issue, was soon discovered to be the real strength of SLAM methods. The analysis is completed by running five selected state-of-the-art SLAM methods, which have been chosen to represent the diversity of existing SLAM designs, on two different datasets These methods best address the use case of pedestrian pose estimation in the urban environment. This experimental benchmark is conducted on a renowned public dataset EuRoC [6] and completed with a new visual-inertial dataset, which has been recorded with smart devices held in hand by a pedestrian in the city center of Nantes in France (IRSTV dataset). A detailed analysis of the SLAM results over the selected dataset completes this section
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
