KITWARE INC — Department of Defense SBIR Phase I: A19-118

Description

There are many challenges in safely maneuvering combat vehicles through a battlespace. Path planning requires detailed knowledge of the terrain and any physical barriers as well as the location of dangers, such as landmines or enemy positions. Small UAVs are now convenient for battlespace reconnaissance using visible and thermal cameras, but the raw imagery needs to be quickly converted to 3D environment maps showing physical obstacles and threats. Kitware proposes Obstacle TeleSculptor to address this need. Obstacle TeleSculptor extends our existing open-source TeleSculptor application for 3D scene modeling from UAV-collected imagery. In addition to mapping the environment, Obstacle TeleSculptor will detect and map both physical obstacles that would block ground vehicles as well as threats such as landmines and enemy vehicles or dismounts that should be avoided. Our system will accumulate detected observations from multiple viewpoints in both visible and thermal imagery and use geometric reasoning to localize and filter results. A deep network will further classify each object using all available observations. The combined map of 3D terrain, obstacles, and threats from timely UAV imagery will provide the critical intelligence needed to make informed decisions about battlespace navigation.