AUTONOMOUS NAVIGATION FOR UNMANNED GROUND VEHICLES IN GPS-DENIED ENVIRONMENTS USING CONVOLUTIONAL NEURAL NETWORKS

Loading...
Thumbnail Image
Authors
Romero, Jensen A.
Subjects
controls
3D map
localization
GPS
GNSS
IMU
UGV
unmanned
autonomous
navigation
ground vehicles
multi-vehicle coordination
distributed environment
obstacle detection
drift
GPS-denied environments
robot
wheeled
artificial intelligence
AI
neural network
convolutional
CNN
Advisors
Calusdian, James
Herman, Jessica L.
Date of Issue
2024-09
Date
Publisher
Monterey, CA; Naval Postgraduate School
Language
Abstract
Reliable navigation in GPS-denied environments remains a challenge for autonomous unmanned systems, as GPS is susceptible to jamming and loss of coverage indoors. This research investigates addressing this challenge for unmanned ground vehicles (UGVs) by integrating convolutional neural networks (CNNs) with visual sensors to achieve real-time pose estimation without GPS reliance. A dual CNN architecture for position and heading estimation was implemented and trained on a substantial dataset of images with corresponding poses. Periodic drift estimation and correction were enabled through integration with a modified potential field algorithm. A key contribution is the world-representation adjustment method for drift correction, by which waypoint positions are dynamically adjusted based on CNN estimates. Successful navigation for multiple consecutive laps in a controlled environment was achieved using this approach, significantly improving upon baseline performance without drift correction. The findings of this research suggest that GPS dependence in autonomous navigation systems could be significantly reduced through this approach, potentially improving the resilience of unmanned systems against electronic warfare tactics and enabling sustained operations in contested environments.
Type
Thesis
Description
Series/Report No
Organization
Identifiers
NPS Report Number
Sponsors
Funder
Format
Citation
Distribution Statement
Distribution Statement A. Approved for public release: Distribution is unlimited.
Rights
This publication is a work of the U.S. Government as defined in Title 17, United States Code, Section 101. Copyright protection is not available for this work in the United States.
Collections