Asymmetric Collaborative Counter Swarm

This federal solicitation (ARM26BX04-NV008) seeks to develop and integrate decentralized, distributed Artificial Intelligence (AI) software to command and control a friendly, multi-agent Group 1 or Group 2 Unmanned Aerial System (UAS) swarm. The system's primary objective is to defend an area by collaborating to counter and degrade a numerically superior enemy swarm in 1-vs-N scenarios. The proposed AI technology must interface with existing platform autonomy and perception systems without relying on a centralized control node. Key technical requirements include operating within extreme Size, Weight, Power, and Cost (SWaP-C) constraints (under a 2 lb payload) and demonstrating the ability to actively collaborate in permissive network environments, as well as passively collaborate with degraded performance in denied or non-permissive environments. Additionally, individual friendly platforms must be capable of degrading or destroying multiple enemy units within a 10-meter range using kinetic or non-kinetic effectors. Phase I offers up to $300,000 for a 1-to-6-month feasibility study to assess algorithmic and sensor configurations. Phase II focuses on developing and demonstrating a prototype system on a Program of Record or custom-built UAS, utilizing hardware for small swarms and simulation for larger scales. Phase III applications include domestic dual-use protection of critical infrastructure. This effort requires a CMMC Level 1 certification.