A Comparison of Modern General-Purpose Visual SLAM Approaches

Advancing maturity in mobile and legged robotics technologies is changing the landscapes where robots are being deployed and found. This innovation calls for a transformation in simultaneous localization and mapping (SLAM) systems to support this new generation of service and consumer robots. No longer can traditionally robust 2D lidar systems dominate while robots are being deployed in multi-story indoor, outdoor unstructured, and urban domains with increasingly inexpensive stereo and RGB-D cameras. Visual SLAM (VSLAM) systems have been a topic of study for decades and a small number of openly available implementations have stood out: ORB-SLAM3, OpenVSLAM and RTABMap.This paper presents a comparison of these 3 modern, feature rich, and uniquely robust VSLAM techniques that have yet to be benchmarked against each other, using several different datasets spanning multiple domains negotiated by service robots. ORB-SLAM3 and OpenVSLAM each were not compared against at least one of these datasets previously in literature and we provide insight through this lens. This analysis is motivated to find general purpose, feature complete, and multi-domain VSLAM options to support a broad class of robot applications for integration into the new and improved ROS 2 Nav2 System as suitable alternatives to traditional 2D lidar solutions.

I. Introduction

Simultaneous localization and mapping (SLAM) has been a field of study for many decades. From radars and range finders, to cameras and lasers, many modalities of SLAM have been developed to solve the fundamental problem of finding sensor poses in a global representation. The development of 2D laser scanners, while expensive, was a game changer that unlocked the mass deployment of service robots seen today. 2D SLAM and localization techniques have led the way for reliable and computationally efficient positioning in large scale environments for much of the last decade. As robot costs are being driven down to enable large scale fleets, sale to consumers, and application to new problems, the cost of laser scanners has become a significant bottleneck. Further, robots are now being deployed in multi-story construction sites, urban cities, and all-terrain areas not suitable for 2D techniques - regardless of sensor price. Robust and mature approaches relying on cameras and comparatively low cost RGB-D sensors will be crucial for enabling this next wave of robot applications.