This thesis presents the development of an exteroceptive sensor suite for real-time detection and classification of navigational markers on Unmanned Surface Vessels. Three sensors were used to complete this task: a 3D LIDAR and two visible light cameras. First, all LIDAR points were transformed from the sensor’s reference frame to the local frame using a Kalman filter to estimate instantaneous vehicle pose. Next, objects were chosen from the LIDAR data to be classified using either Multivariate Gaussian or Parzen Window Classifiers. Both produce 96% accuracy or better, however, multivariate Gaussian ran considerably faster than the Parzen and was simpler to implement and was therefore chosen as the final classifier. Additionally, regions of interest images based on the Multivariate Gaussian classification were extracted from the full camera images to improve marker knowledge. This sensor suite and set of algorithms underwent extensive testing on Embry-Riddle’s Maritime RobotX and RoboBoat platforms and greatly improves the ability to quickly and accurately identify multiple navigational markers, which is paramount to the success of any Unmanned Surfaces Vessel.
","abstract_html":"<p>This thesis presents the development of an exteroceptive sensor suite for real-time detection and classification of navigational markers on Unmanned Surface Vessels. Three sensors were used to complete this task: a 3D LIDAR and two visible light cameras. First, all LIDAR points were transformed from the sensor’s reference frame to the local frame using a Kalman filter to estimate instantaneous vehicle pose. Next, objects were chosen from the LIDAR data to be classified using either Multivariate Gaussian or Parzen Window Classifiers. Both produce 96% accuracy or better, however, multivariate Gaussian ran considerably faster than the Parzen and was simpler to implement and was therefore chosen as the final classifier. Additionally, regions of interest images based on the Multivariate Gaussian classification were extracted from the full camera images to improve marker knowledge. This sensor suite and set of algorithms underwent extensive testing on Embry-Riddle’s Maritime RobotX and RoboBoat platforms and greatly improves the ability to quickly and accurately identify multiple navigational markers, which is paramount to the success of any Unmanned Surfaces Vessel.</p>","abstract_has_math":false,"creators":["Kennedy, Christopher Lloyd"],"institution":null,"degree_name":"Master of Science in Mechanical Engineering","degree_level":"Thesis - Open Access","degree_discipline":"Mechanical Engineering","degree_department":null,"school":null,"contributors":[],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2014,"date_issued":"2014-12-01T08:00:00Z","date_published":"2014-12-01T08:00:00Z","updated_at":"2026-07-27T19:26:28Z","subjects":["exteroceptive","sensors","unmanned surface vehicles","navigational markers","Aeronautical Vehicles","Mechanical Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://commons.erau.edu/edt/167","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:creator","label":"Author","values":["Kennedy, Christopher Lloyd"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"thesis:degree_discipline","label":"Discipline","values":["Mechanical Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis - Open Access"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Mechanical Engineering"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["exteroceptive","sensors","unmanned surface vehicles","navigational markers","Aeronautical Vehicles","Mechanical Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://commons.erau.edu/edt/167"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis presents the development of an exteroceptive sensor suite for real-time detection and classification of navigational markers on Unmanned Surface Vessels. Three sensors were used to complete this task: a 3D LIDAR and two visible light cameras. First, all LIDAR points were transformed from the sensor’s reference frame to the local frame using a Kalman filter to estimate instantaneous vehicle pose. Next, objects were chosen from the LIDAR data to be classified using either Multivariate Gaussian or Parzen Window Classifiers. Both produce 96% accuracy or better, however, multivariate Gaussian ran considerably faster than the Parzen and was simpler to implement and was therefore chosen as the final classifier. Additionally, regions of interest images based on the Multivariate Gaussian classification were extracted from the full camera images to improve marker knowledge. This sensor suite and set of algorithms underwent extensive testing on Embry-Riddle’s Maritime RobotX and RoboBoat platforms and greatly improves the ability to quickly and accurately identify multiple navigational markers, which is paramount to the success of any Unmanned Surfaces Vessel.
"]},{"key":"dc:title","label":"Title","values":["Development of an Exteroceptive Sensor Suite on Unmanned Surface Vessels for Real-Time Classification of Navigational Markers"]}]}],"canonical_facts":{"dc:creator":["Kennedy, Christopher Lloyd"],"dc:description.abstract":["This thesis presents the development of an exteroceptive sensor suite for real-time detection and classification of navigational markers on Unmanned Surface Vessels. Three sensors were used to complete this task: a 3D LIDAR and two visible light cameras. First, all LIDAR points were transformed from the sensor’s reference frame to the local frame using a Kalman filter to estimate instantaneous vehicle pose. Next, objects were chosen from the LIDAR data to be classified using either Multivariate Gaussian or Parzen Window Classifiers. Both produce 96% accuracy or better, however, multivariate Gaussian ran considerably faster than the Parzen and was simpler to implement and was therefore chosen as the final classifier. Additionally, regions of interest images based on the Multivariate Gaussian classification were extracted from the full camera images to improve marker knowledge. This sensor suite and set of algorithms underwent extensive testing on Embry-Riddle’s Maritime RobotX and RoboBoat platforms and greatly improves the ability to quickly and accurately identify multiple navigational markers, which is paramount to the success of any Unmanned Surfaces Vessel.
"],"dc:identifier":["https://commons.erau.edu/edt/167"],"dc:subject":["exteroceptive","sensors","unmanned surface vehicles","navigational markers","Aeronautical Vehicles","Mechanical Engineering"],"dc:title":["Development of an Exteroceptive Sensor Suite on Unmanned Surface Vessels for Real-Time Classification of Navigational Markers"],"thesis:degree_discipline":["Mechanical Engineering"],"thesis:degree_level":["Thesis - Open Access"],"thesis:degree_name":["Master of Science in Mechanical Engineering"]},"updated_at":"2026-07-27T19:26:28Z"}