Long-Range Thermal 3D Perception in Low Contrast Environments

The proposed technology intends to improve the future air vehicles#39; situational awareness by providing a novel 3D perception capability targeting the critical for the AAM applications gap between currently available short-range automotive LiDAR and long-range radars with the limited transversal and temporal resolution, especially for the obstacle-rich urban environment. It will provide high-contrast day/night and all-weather image data, independent of GPS that allows robust depth sensing at hundreds-to-thousands of meters range and high vehicle speed. This robustness of depth perception contributes to the overall safety of NAS operations.We propose a multi-sensor long-wave infrared (LWIR) 3D technology that mitigates the limitations of the compact uncooled thermal sensors: low contrast, low resolution, and high thermal inertia. The novel sensor configuration, calibration methods, and image processing algorithms yields composite system performance by far exceeding that of the individual COTS thermal imagers#39; performance. Using the LWIR sensors with noise-equivalent temperature difference (NETD)=40 mK, this technology will demonstrate system-equivalent NETD of 10 mK in the intrascene mode and 1 mK for the low-contrast environmental features, by order of magnitude exceeding the performance of cryogenic photon detectors in the traditional binocular configuration.This proposal#39;s research objective is to build the LWIR/EO multi-sensor prototype and demonstrate the dramatic improvement of the microbolometer-based LWIR detectors sensitivity, especially for the 3D measurements and motion blur mitigation. The results of this research will enable passive thermal 3D perception systems for a broad spectrum of environmental conditions. It will pave the way to using large arrays of low SWaP-C microbolometer devices advanced by consumer thermal imaging products.