LOCALIZATION PERFORMANCE ESTIMATION WITH MULTI SENSOR FUSION

Download

Ulas_Thesis.pdf

Date

2022-5-12

Author

Bingöl, Ulaş Süreyya

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

350
views

251
downloads

Autonomous navigation gained significant attention as it is necessary for an over- growing use and integration of autonomous systems to numerous fields such as self- driving cars, logistics, and agricultural robots. These applications require robots to navigate from one location to another, which requires them to estimate their position accurately. Localization loss scenarios are instances when localization accuracy dete- riorates to a level that endangers the task. Reaching the false end goals, being stuck, traffic jams, and even crashes can occur when robots perform navigation with dete- riorated localization. Specifically for autonomous agents working alongside humans or as a fleet, localization accuracy is crucial. This work presents a novel method for monitoring and estimating localization per- formance of individual agents, called Localization Performance Analyzer (LPA). In contrast to the previous studies in this field, rather than specifically focusing on a few metrics to capture this deterioration, we instead investigate several metrics to find out which are more crucial for the estimation of localization failures. The algorithm first extracts a set of features from the robots’ sensors and its pose belief, and feeds them to a neural network (NN). Then, we train this network with the data from multiple robots, and able to predict whether the robot is lost or localized. We designed LPA in order to increase autonomy and to recover from localization divergences of a fleet of autonomous mobile robots (AMR). The effectiveness of the proposed algorithm is validated with the data gathered from industrial AMRs working in the same field.

Subject Keywords

Localization, Autonomous Mobile Robots, Autonomy

URI

https://hdl.handle.net/11511/97342

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Trajectory planning and tracking for autonomous vehicles Çiçek, Haluk Levent; Schmidt, Klaus Verner; Department of Electrical and Electronics Engineering (2022-12-27) Finding appropriate paths is an essential issue for the development of autonomous vehicles and robots. Hereby, it has to be considered that autonomous vehicles cannot follow sharp corners, as they cannot turn on a single point. Therefore, it is important to compute smooth paths that have additional desirable properties such as minimum length and sufficient distance from obstacles. Furthermore, practical applications require the computation of such paths in real time. This thesis develops a general method...
Feedback Motion Planning For a Dynamic Car Model via Random Sequential Composition Özcan, Melih; Ankaralı, Mustafa Mert (2019-01-01) Autonomous cars and car-like robots have gained huge popularity recently due to the recent advancements in technology and AI industry. Motion and path planning is one of the most fundamental problems for such systems. In the literature, kinematic models are widely adopted for planning and control for these type of robots due to their simplicity (control and analysis) and fewer computational requirements. Though, applicability of kinematic models are limited to very low speeds or some specific cases, which c...
Communication and coordination for urban intelligent transportation: architecture and algorithms Atagoziev, Maksat; Schmidt, Klaus Verner; Schmidt, Şenan Ece; Department of Electrical and Electronics Engineering (2022-2-10) In the scope of Intelligent Transportation Systems (ITS), the automation and coordination of connected and autonomous vehicle (CAV) lane changes (LCs) have a strong impact on driving safety and traffic throughput. Accordingly, this thesis develops algorithms for the coordination of CAV LCs that are then used for controlling the traffic at intersections. First, this thesis focuses on the coordination of LCs of a group of CAVs to minimize the time when all LCs are completed, while keeping small inter-vehicle ...
Localization for mobile robots using skyline in panoramic images Koku, Ahmet Buğra (Journal of the Faculty of Engineering and Architecture of Gazi University, 2019-01-01) Advances in robotics liberated robots from factory floors by the end of 20th century. Use of robots in our daily lives is only expected to increase in time. Robots, while relieving us of the burden of tedious, hard and dangerous tasks, most of them are still expected to be territorial, i.e. they will operate in a predefined or rather bounded environment. For enhanced performance, a robot should be familiar with its territory. In this work, use of skylines extracted from panoramic images is studied in order ...
Unstructured road recognition and following for mobile robots via image processing using Anns Dilan, Rasim Aşkın; Koku, Ahmet Buğra; Department of Mechanical Engineering (2010) For an autonomous outdoor mobile robot ability to detect roads existing around is a vital capability. Unstructured roads are among the toughest challenges for a mobile robot both in terms of detection and navigation. Even though mobile robots use various sensors to interact with their environment, being a comparatively low-cost and rich source of information, potential of cameras should be fully utilized. This research aims to systematically investigate the potential use of streaming camera images in detect...

Citation Formats

U. S. Bingöl, “LOCALIZATION PERFORMANCE ESTIMATION WITH MULTI SENSOR FUSION,” M.S. - Master of Science, Middle East Technical University, 2022.