3D Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR) in Contested Environments

OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Trusted AI and Autonomy; Integrated Sensing and Cyber; Integrated Network System-of-Systems The technology within this topic is restricted under the International Traffic in Arms Regulation (ITAR), 22 CFR Parts 120-130, which controls the export and import of defense-related material and services, including export of sensitive technical data, or the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) in accordance with the Announcement. Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US Export Control Laws. OBJECTIVE: The objective is to develop solutions that empower the ability to create 3D Synthetic Aperture Radar (SAR) images and accurately identify multiple class of ground and surface targets using an Automatic Target Recognition (ATR) algorithm. Contested environments can represent any domain where 3D target information is not able to be collected through long loiter or persistent access. Potential solution(s) include employing an airborne and/or spaceborne platform of various classes within the solution space. The solution (threshold) of detecting, tracking, geolocating and 3D imaging with sufficient quality (objective) to enable classifying and identifying targets for ingestion in a 3D ATR algorithm. DESCRIPTION: Presently, automatic target recognition (ATR) solutions have been extensively studied in the two-dimensional (2D) cases, given the ease of data representation. However, 2D images are susceptible to ambiguity in the target representation, given the dimensionality reduction from representing objects that humans perceive in three-dimensions (3D) in a 2D format. Current 3D electro-optical (EO) ATR techniques have been shown to perform better than their 2D EO ATR counterparts, by removing projection artifacts that are inherently seen in 2D ATRs. However, EO-based ATRs are susceptible to low-lighting/weather conditions that may not be feasible in contested environments. SAR's ability to operate in all-weather/illumination scenarios provides a natural mitigation to these conditions and is thus the primary focus of this topic. State of the art SAR ATR currently addresses 2D scenarios, not currently scaling up to 3D scenarios. Similarly, current 3D SAR imaging-based solutions have not yet begun exploring ATR with their reconstructions. This topic aims to investigate approaches for 3D SAR ATR that are suitable with 3D SAR imagery of targets in contested airborne and/or spaceborne environments. PHASE I: Awardee(s) exepect to provide a detailed technical approach to achieving this goal by stating platforms and configurations for achieving 3D imagery within the contested environment and the approach for testing and training the 3D ATR algorithm for correct classification. The awardee(s) will describe the constraints of the collection, including necessary operating conditions, platforms, and algorithms to achieve 3D image formation. The approach should provide solutions for ground and littoral settings and provide any constraints for application across various target types and platforms. PHASE II: Awardee(s) will implement the algorithmic approach outlined in PHASE I. Basic processing requirements are detection, short time tracking within a coherent dwell (generally less than 10 seconds), motion stabilization, geolocation, imaging, and executing a 3D SAR ATR. Efficiency of algorithms is important to make use of low cost, size, weight and power (low C-SWAP) platforms. Offerors are expected to generate synthetic data to create proof of concept algorithms as part of the effort. The government may supply additional measured supplemental data to validate synthetic performance results. The awardee(s) will train their algorithms and provide metrics of success such as probability of correct identification (PID), probability of detection (PDet), and probability of correct classification (PCC). PHASE III DUAL USE APPLICATIONS: The algorithm for the 3D ATR approach will be implemented on a representative platform and tested against targets of classes of interest. Transition customers will be identified and integration and interoperability requirements will be determined for fielding of the algorithm. If a viable business model for the developed strategy or algorithm(s) is demonstrated, the offeror or identified transition partners could be in a position to supply future processes to the Air Force and other DoD components as this process is adopted. REFERENCES: 1. Nevis, Andrew J., et al. "Advantages of three-dimensional electro-optic imaging sensors." Detection and Remediation Technologies for Mines and Minelike Targets VIII. Vol. 5089. SPIE, 2003. 2. Nguyen, Lam H., and Calvin Le. "3D imaging for millimeter-wave forward-looking synthetic aperture radar (SAR)." Passive and Active Millimeter-Wave Imaging XXIII. Vol. 11411. SPIE, 2020. 3. Sugavanam, Nithin, Emre Ertin, and Jan Rainer Jamora. "Deep learning for three dimensional SAR imaging from limited viewing angles." 2023 IEEE International Radar Conference (RADAR). IEEE, 2023. 4. Green, Dylan, J. R. Jamora, and Anne Gelb. "Leveraging joint sparsity in 3D synthetic aperture radar imaging." Applied Mathematics for Modern Challenges 1.1 (2023): 61-86. KEYWORDS: 3D SAR ATR; automatic target recognition; three dimensional; synthetic aperture radar; SAR ATR; 3D ATR