Survey on Localization Systems and Algorithms for Unmanned Systems

Abstract

Intelligent unmanned systems have important applications, such as pesticide-spraying in agriculture, robot-based warehouse management systems, and missile-firing drones. The underlying assumption behind all autonomy is that the agent knows its relative position or egomotion with respect to some reference or scene. There exist thousands of localization systems in the literature. These localization systems use various combinations of sensors and algorithms, such as visual/visual-inertial SLAM, to achieve robust localization. The majority of the methods use one or more sensors from LIDAR, camera, IMU, UWB, GPS, compass, tracking system, etc. This survey presents a systematic review and analysis of published algorithms and techniques chronologically, and we introduce various highly impactful works. We provide insightful investigation and taxonomy on sensory data forming principle, feature association principle, egomotion estimation formation, and fusion model for each type of system. At last, some open problems and directions for future research are also included. We aim to survey the literature comprehensively to provide a complete understanding of localization methodologies, performance, advantages and limitations, and evaluations of various methods, shedding some light for future research.

Towards Real-time Semantic RGB-D SLAM in Dynamic Environments

“…Because a single object could be segmented into a few clusters, the object is not required to be rigid, while most semantic SLAM methods have this rigidity assumption. For each cluster c j , we compute an average reprojection error r j of all feature points u i inside the cluster against their matched correspondences P i in 3D space, as defined in (1), where m is the number of matched features in c j , T wc is the camera pose, π represents the camera projection model, and ρ is a penalty function.…”

“…Simultaneous Localization and Mapping (SLAM) is a fundamental capability for intelligent robotic applications. It aims to simultaneously estimate the poses of the robot and build a map of the unknown environment from the data of onboard sensors [1], [2]. Benefiting from various low-cost and light-weight cameras available in the market, vision-based SLAM, or visual SLAM (vSLAM), has received increasing attention over the past decades.…”

“…Although robust estimation techniques such as RANSAC can be applied to filter out some of the outliers, the improvement is still limited since they can This work was partly supported by the Delta-NTU Corporate Laboratory for Cyber-Physical Systems funded by the National Research Foundation (NRF) Singapore under its Corporate Laboratory@University Scheme. 1 Tete Ji and Lihua Xie are with the School of Electrical and Electronic Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798. {jitete, elhxie}@ntu.edu.sg 2 Chen Wang is with the Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA.…”

A Novel Multispectral Maritime Target classification based on ThermalGAN (RGB-to-Thermal Image Translation)

“…Due to the large number of maritime images to be classified, an artificial intelligence approach based on deep learning (Mclean et al, 2021;Su et al, 2020) is very important to extract and classify different images automatically (Yuan et al, 2021). We considered three CNN architectures for ship classification.…”

A Survey of Robot Swarms’ Relative Localization Method

“…As a detailed review of positioning, ref. [12,28,29] summarized the working principle of the existing positioning system and some challenges existing in the positioning system, most of which are aimed at a single robot, and few or no solutions are proposed. Therefore, it is crucial to summarize and discuss the localization method for robot swarms and determine the direction of further research.…”