A Viewpoint Selection-based Coverage Planning Algorithm for Autonomous Structural Inspection using Multiple UAVs
Advisor Information
Dr. Prithviraj Dasgupta
Location
UNO Criss Library, Room 225
Presentation Type
Oral Presentation
Start Date
2-3-2018 11:00 AM
End Date
2-3-2018 11:15 AM
Abstract
Autonomous Unmanned Aerial Vehicles (UAVs) are being used extensively in structural inspection tasks to inspect bridges, buildings and other structures. The main objective of a UAV in a structural inspection task is to completely cover the exposed surface of the structure, while collecting information from the surface through its onboard sensors including camera, LIDAR, laser, etc. Using a single UAV for autonomously covering a large 3D structure such as a bridge might not be feasible owing to battery limitations of the UAV. To address this problem, we propose a multi-UAV, viewpoint-based approach, where, first, a set of points or locations around the structure is identified so that the entire structure can be covered while visiting these locations. Subsequently, the locations are allocated among a team of UAVs so that each UAV visits a subset of the locations while ensuring that the UAVs do not collide with each other and the effort (energy) expended by the UAVs for visiting the locations is balanced across the UAVs.
A Viewpoint Selection-based Coverage Planning Algorithm for Autonomous Structural Inspection using Multiple UAVs
UNO Criss Library, Room 225
Autonomous Unmanned Aerial Vehicles (UAVs) are being used extensively in structural inspection tasks to inspect bridges, buildings and other structures. The main objective of a UAV in a structural inspection task is to completely cover the exposed surface of the structure, while collecting information from the surface through its onboard sensors including camera, LIDAR, laser, etc. Using a single UAV for autonomously covering a large 3D structure such as a bridge might not be feasible owing to battery limitations of the UAV. To address this problem, we propose a multi-UAV, viewpoint-based approach, where, first, a set of points or locations around the structure is identified so that the entire structure can be covered while visiting these locations. Subsequently, the locations are allocated among a team of UAVs so that each UAV visits a subset of the locations while ensuring that the UAVs do not collide with each other and the effort (energy) expended by the UAVs for visiting the locations is balanced across the UAVs.