{"id":{"repo_id":"nps","oai_identifier":"oai:calhoun.nps.edu:10945/74230"},"canonical_url":"https://search.dev.ndltd.org/etd/nps/oai:calhoun.nps.edu:10945/74230","repository":{"repo_id":"nps","name":"Naval Postgraduate School","base_url":"https://calhoun.nps.edu/server/oai/request"},"display":{"title":"A RAPID PROTOTYPING FRAMEWORK FOR VISION-BASED LOCALIZATION ALGORITHMS","abstract":"Unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs) traditionally rely on the Global Positioning System (GPS). As adversarial electronic warfare capabilities grow, GPS signals face ever-increasing risks of jamming and spoofing, especially in contested environments. Therefore, as these adversarial risks increase, alternatives to GPS have become necessary. Vision-based navigation and localization algorithms, which estimate positional data using an onboard camera sensor, offer a promising solution. To accelerate the development of such an approach, a simulation tool is necessary for rapidly prototyping and evaluating these algorithms before implementation in a realistic environment. This research introduced an open-source simulation framework developed using the Robot Operating System (ROS) and Gazebo. Within the simulation, a UAV equipped with an onboard camera and a YOLO11 convolutional neural network (CNN) detected and tracked both a friendly and an enemy vessel. A particle filter then estimated relative positional data, enabling decision-making without GPS. The framework supported plug-and-play integration and modification of estimators, object detection models, and sensor parameters, allowing for future development and expansion. This work demonstrated the feasibility and various challenges of vision-based localization, laying a foundation for real-world testing of autonomous systems operating in GPS-denied environments.","abstract_html":"Unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs) traditionally rely on the Global Positioning System (GPS). As adversarial electronic warfare capabilities grow, GPS signals face ever-increasing risks of jamming and spoofing, especially in contested environments. Therefore, as these adversarial risks increase, alternatives to GPS have become necessary. Vision-based navigation and localization algorithms, which estimate positional data using an onboard camera sensor, offer a promising solution. To accelerate the development of such an approach, a simulation tool is necessary for rapidly prototyping and evaluating these algorithms before implementation in a realistic environment. This research introduced an open-source simulation framework developed using the Robot Operating System (ROS) and Gazebo. Within the simulation, a UAV equipped with an onboard camera and a YOLO11 convolutional neural network (CNN) detected and tracked both a friendly and an enemy vessel. A particle filter then estimated relative positional data, enabling decision-making without GPS. The framework supported plug-and-play integration and modification of estimators, object detection models, and sensor parameters, allowing for future development and expansion. This work demonstrated the feasibility and various challenges of vision-based localization, laying a foundation for real-world testing of autonomous systems operating in GPS-denied environments.","abstract_has_math":false,"creators":["Zukowski, Braden E."],"institution":"Monterey, CA; Naval Postgraduate School","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":"Mechanical and Aerospace Engineering (MAE)","school":null,"contributors":[],"advisors":["Kragelund, Sean P."],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-06","date_published":"2025-06","updated_at":"2026-07-27T20:26:50Z","subjects":[],"languages":[],"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."],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/10945/74230","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Kragelund, Sean P."]},{"key":"dc:contributor.department","label":"Department","values":["Mechanical and Aerospace Engineering (MAE)"]},{"key":"dc:creator","label":"Author","values":["Zukowski, Braden E."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-09-08T15:59:36Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-09-08T15:59:36Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-06"]},{"key":"dc:publisher","label":"Institution","values":["Monterey, CA; Naval Postgraduate School"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights","label":"Dc Rights","values":["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."]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10945/74230"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs) traditionally rely on the Global Positioning System (GPS). As adversarial electronic warfare capabilities grow, GPS signals face ever-increasing risks of jamming and spoofing, especially in contested environments. Therefore, as these adversarial risks increase, alternatives to GPS have become necessary. Vision-based navigation and localization algorithms, which estimate positional data using an onboard camera sensor, offer a promising solution. To accelerate the development of such an approach, a simulation tool is necessary for rapidly prototyping and evaluating these algorithms before implementation in a realistic environment. This research introduced an open-source simulation framework developed using the Robot Operating System (ROS) and Gazebo. Within the simulation, a UAV equipped with an onboard camera and a YOLO11 convolutional neural network (CNN) detected and tracked both a friendly and an enemy vessel. A particle filter then estimated relative positional data, enabling decision-making without GPS. The framework supported plug-and-play integration and modification of estimators, object detection models, and sensor parameters, allowing for future development and expansion. This work demonstrated the feasibility and various challenges of vision-based localization, laying a foundation for real-world testing of autonomous systems operating in GPS-denied environments."]},{"key":"dc:title","label":"Title","values":["A RAPID PROTOTYPING FRAMEWORK FOR VISION-BASED LOCALIZATION ALGORITHMS"]}]}],"canonical_facts":{"dc:contributor.advisor":["Kragelund, Sean P."],"dc:contributor.department":["Mechanical and Aerospace Engineering (MAE)"],"dc:creator":["Zukowski, Braden E."],"dc:date.accessioned":["2025-09-08T15:59:36Z"],"dc:date.available":["2025-09-08T15:59:36Z"],"dc:date.issued":["2025-06"],"dc:description.abstract":["Unmanned aerial vehicles (UAVs) and unmanned surface vessels (USVs) traditionally rely on the Global Positioning System (GPS). As adversarial electronic warfare capabilities grow, GPS signals face ever-increasing risks of jamming and spoofing, especially in contested environments. Therefore, as these adversarial risks increase, alternatives to GPS have become necessary. Vision-based navigation and localization algorithms, which estimate positional data using an onboard camera sensor, offer a promising solution. To accelerate the development of such an approach, a simulation tool is necessary for rapidly prototyping and evaluating these algorithms before implementation in a realistic environment. This research introduced an open-source simulation framework developed using the Robot Operating System (ROS) and Gazebo. Within the simulation, a UAV equipped with an onboard camera and a YOLO11 convolutional neural network (CNN) detected and tracked both a friendly and an enemy vessel. A particle filter then estimated relative positional data, enabling decision-making without GPS. The framework supported plug-and-play integration and modification of estimators, object detection models, and sensor parameters, allowing for future development and expansion. This work demonstrated the feasibility and various challenges of vision-based localization, laying a foundation for real-world testing of autonomous systems operating in GPS-denied environments."],"dc:identifier.uri":["https://hdl.handle.net/10945/74230"],"dc:publisher":["Monterey, CA; Naval Postgraduate School"],"dc: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."],"dc:title":["A RAPID PROTOTYPING FRAMEWORK FOR VISION-BASED LOCALIZATION ALGORITHMS"],"dc:type":["Thesis"]},"updated_at":"2026-07-27T20:26:50Z"}