Abstract
Self-localization enables a system to navigate and interact with its environment. In this study, we propose a novel sparse semantic self-localization approach for robust and efficient indoor localization. “Sparse semantic” refers to the detection of sparsely distributed objects such as doors and windows. We use sparse semantic information to self-localize on a human-readable 2D annotated map in the sensor model. Thus, compared to previous works using point clouds or other dense and large data structures, our work uses a small amount of sparse semantic information, which efficiently reduces uncertainty in real-time localization. Unlike complex 3D constructions, the annotated map required by our method can be easily prepared by marking the approximate centers of the annotated objects on a 2D map. Our approach is robust to the partial obstruction of views and geometrical errors on the map. The localization is performed using low-cost lightweight sensors, an inertial measurement unit and a spherical camera. We conducted experiments to show the feasibility and robustness of our approach.
Highlights
Self-localization—i.e., obtaining one’s pose information within a map—is necessary to navigate and interact with the environment
We propose a sparse semantic localization method with the following features
Instead of using distance measurements, Look-Up Table (LUT) or the combination of likelihood and beam models, we propose the use of only a relative bearing angle to detect the object center and the object class names as sensor inputs; we propose that each object class should be treated as a different sensor with different weights for indoor localization
Summary
Self-localization—i.e., obtaining one’s pose information within a map—is necessary to navigate and interact with the environment. Localization methods have a wide range of applications, from monitoring robots to facilitating assistance systems. Depending on the system, these methods face challenges in terms of factors such as achieving acceptable accuracy and robustness as well as reducing complexity and sensor costs. The outdoor localization issue is mostly solved using GPS. Because GPS is not very effective inside buildings, indoor localization remains a challenge. In the absence of GPS systems, different information sources are required for indoor localization. Radio frequency ID (RFID) tags, ultrasonic tags and infrared emitters are potentially viable alternatives; these methods require prior installation and modifications to the infrastructure [1,2,3]
Sign-in/Register to view highlights & summary 
Published Version (
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