Iterative Tomographic Algorithm for Laser Radar Imaging of Unmanned-Aerial Vehicles

Unmanned-Aerial Vehicles (UAVs) present a unique threat to deployed ground forces overseas and civilian population centers in the US. Phase II will continue the development of a ladar-based imaging and surveillance system for perimeter defense against UAVs. In the successful Phase I, a characterization of two-dimensional limited-angle ladar-UAV imaging was performed. The system, Tomographic Reconstruction of Active Imaging Ladar Signals (TRAILS), will be upgraded to include three-dimensional limited-angle tomographic reconstruction, high fidelity forward simulation of ladar reflection from UAVs, atmospheric effects, target identification, and ladar specification analysis. This includes SNR and image resolution tradeoffs under various imaging conditions, including severely occluded imaging sweeps and UAV self-shadowing. Forefront SSI codes for light and atmospheric simulation, QUID and MODTRAN, will be interfaced with limited-angle 3D reconstruction algorithms. The innovative use of OpenGL buffers for the reconstruction algorithms will make the reconstruction near real time. Spectral Sciences has teamed with Kodak to measure plume absorption profiles from an airborne differential absorption lidar (DIAL) sensor, and apply TRAILS to reconstruct the plume in three dimensions. In addition to testing TRAILS for UAV threat scenarios, the collaboration will lead to new applications for TRAILS.