Autonomous Unmanned Aircraft Systems (UAS) Intelligence, Surveillance, Reconnaissance and Targeting (ISR-T)

OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Trusted AI and Autonomy; Advanced Computing and Software 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 USAF is actively seeking a low Size, Weight, and Power (SWaP) solution for a group 2 or 3 UAS to conduct Intelligence, Surveillance, Reconnaissance and Targeting (ISR-T) in highly congested and contested operational environments. The low-SWaP solution should allow the UAS to autonomously acquire and track specific ground targets using EO/IR video and RF sensors. The UAS will conduct its mission and exit the area with limited or no communications while being jammed or spoofed. To avoid detection, the ground targets may employ Camouflage, Concealment, and Deception (CCD) or may hide among many other vehicles (e.g., a parking lot) [1]. Alternatively, to increase tracking difficulty, the ground targets may attempt evasive motion and/or high-speed turns. DESCRIPTION: Modern UAS designs can operate autonomously, avoiding collisions with other UAS, buildings, and terrain using low power/short-range sensors. A ground target attempting to evade the UAS may use occlusions (tunnels, bridges, foliage, parking garage etc.) to temporarily hide from the surveillance. Through the combination of a gimballed camera system and high camera frame rate, the UAS can detect a partially hidden target, track a ground target robustly through aggressive maneuvers, and also recognize a particular target vehicle when surrounded by many similar vehicles. A group 2 or 3 UAS can autonomously search an area using a flight pattern at low altitude for camouflaged or concealed ground targets without GPS, command and control (C2), or video signals (so that it cannot be jammed or spoofed) [2, 3]. To accomplish this task, the numbers of pixels per target must be sufficient for the Automated Target Recognition (ATR) algorithms despite blur [4]. The UAS may initially detect a concealed ground vehicle using an RF signal (e.g., communications, radar, or unintentional). PHASE I: Implement the EO/IR and RF signal processing, flight optimization, and target detection/tracking algorithms via real-time software. Demonstrate queuing and tracking detections in realistic, high-fidelity coupled EO/IR and RF simulations. The RF simulations must incorporate site-specific scattering and multipath [5]. Transfer sections of the real-time software to a low-SWaP processing solution. Validate performance of the processing algorithms and low-SWaP hardware in terms of tracking accuracy and power consumption for simulated flight tests in realistic environments. PHASE II: In development flight tests, demonstrate autonomous single or cooperative UAS swarm detection of ground vehicle classes using RF signals of opportunity and/or EO/IR video. Establish tracking performance for specific ground vehicles through aggressive maneuvers, temporary occlusions, and CCD. UAS will approach vehicles from behind to avoid detection. UAS will not utilize GPS or C2 signals in contested areas. PHASE III DUAL USE APPLICATIONS: The proposer will work with the Air Force to identify military platforms that could benefit from this technology and develop plans for integration. Commercial applications of UASs are growing significantly year over year. Law enforcement, prison protection, entertainment, package delivery in challenging environments, etc. In many of these applications, there is a need for a UAS to detect, ID and track a variety of objects such as ground vehicles and even people, etc. The proposed topic will contribute directly to these commercial dual use applications. REFERENCES: 1. R. Visina, Y. Bar-Shalom, and P. Willett, "Multiple-model estimators for tracking sharply maneuvering ground targets," IEEE Transactions on Aerospace and Electronic Systems, vol. 54, no. 3, pp. 1404-1414, 2018. 2. C. Wang, J. Wang, X. Zhang, and X. Zhang, "Autonomous navigation of UAV in large-scale unknown complex environment with deep reinforcement learning," in 2017 IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2017, pp. 858-862: Ieee. 3. G. Ciaparrone, F. L. S nchez, S. Tabik, L. Troiano, R. Tagliaferri, and F. Herrera, "Deep learning in video multi-object tracking: A survey," Neurocomputing, vol. 381, pp. 61-88, 2020. 4. T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Doll r, "Focal loss for dense object detection," in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2980-2988. 5. S. Gogineni et al., "High fidelity RF clutter modeling and simulation," IEEE Aerospace and Electronic Systems Magazine, vol. 37, no. 11, pp. 24-43, 2022. KEYWORDS: Target Tracking; Sensor Fusion; Automated Target Recognition (ATR); GPS-Denied Environments; Flight Optimization; Deep Learning Neural Network; Neuromorphic Computing