VPS-SLAM: Visual Planar Semantic SLAM for Aerial Robotic Systems

Abstract

Indoor environments have abundant presence of high-level semantic information which can provide a better understanding of the environment for robots to improve the uncertainty in their pose estimate. Although semantic information has proved to be useful, there are several challenges faced by the research community to accurately perceive, extract and utilize such semantic information from the environment. In order to address these challenges, in this paper we present a lightweight and real-time visual semantic SLAM framework running on board aerial robotic platforms. This novel method combines low-level visual/visual-inertial odometry (VO/VIO) along with geometrical information corresponding to planar surfaces extracted from detected semantic objects. Extracting the planar surfaces from selected semantic objects provides enhanced robustness and makes it possible to precisely improve the metric estimates rapidly, simultaneously generalizing to several object instances irrespective of their shape and size. Our graph-based approach can integrate several state of the art VO/VIO algorithms along with the state of the art object detectors in order to estimate the complete 6DoF pose of the robot while simultaneously creating a sparse semantic map of the environment. No prior knowledge of the objects is required, which is a significant advantage over other works. We test our approach on a standard RGB-D dataset comparing its performance with the state of the art SLAM algorithms. We also perform several challenging indoor experiments validating our approach in presence of distinct environmental conditions and furthermore test it on board an aerial robot.

Active Mapping and Robot Exploration: A Survey

“…Having semantic attributes distinction among objects instead of only geometric entities is necessary so the robot can understand the scene surrounding it. This capability has brought big improvements in classical SLAM [ 145 , 146 , 147 ] and can be particularly useful to explore unknown regions. Ekvall et al [ 148 ] integrate an object recognition system into SLAM in a service robot scenario.…”

Graph SLAM Built over Point Clouds Matching for Robot Localization in Tunnels

“…Another type of SLAM is vision-based SLAM, which might not operate well in environments with a low quantity of features and illumination changes, like tunnels. Visual SLAM may be feature-based (sparse, semi-dense, or dense) or intensity-based [ 16 ]. Most semi-dense SLAM techniques, like [ 17 , 18 ], rely on low-level characteristic features of the environment, such as corners, points, lines, and planes.…”

A Survey of Computer Vision Methods for 2D Object Detection from Unmanned Aerial Vehicles

“…Semantic situational awareness intends to incorporate higher-level semantic information that augment the basic metric models of the environment, as for example [ 37 ]. Object detection provides a very valuable semantic information that can be exploited when building semantic maps, such as in [ 37 , 38 , 39 , 40 , 41 , 42 ].…”