Autonomous UUV Swarm for Undersea Threat Detection & Response

Develop an advanced autonomous undersea vehicle (UUV) swarm coordination system designed to detect, classify, and track unknown underwater objects in challenging, dynamic environments. This challenge emphasizes multi-sensor fusion, real-time decision-making, and intelligent mission planning to effectively address maritime security threats. The system should utilize heterogeneous sensor data (e.g., acoustic, sonar) to form a coherent understanding of the underwater battlespace. A core component involves integrating a large language model, specifically Llama 3.1, to interpret complex scenarios, assess potential threats, and dynamically re-plan UUV trajectories or search patterns based on observed anomalies and overarching mission objectives. Participants will build a simulated environment to test UUV capabilities under various conditions, including communication constraints and environmental noise. The solution should demonstrate robustness, efficient swarm coordination, and effective threat neutralization strategies, deploying edge ML models using BentoML for on-board processing.

• Master principles of multi-robot coordination and swarm intelligence (e.g., consensus algorithms, decentralized control) relevant to UUV operations, ensuring efficient coverage and target handoff.

• Implement advanced sensor fusion techniques (e.g., Extended Kalman Filters, Particle Filters, Unscented Kalman Filters) to combine heterogeneous simulated sensor data (e.g., passive sonar bearing, active sonar range-bearing, visual/Lidar in clear water).

• Design and build a simulated UUV environment (e.g., using Gazebo, Unity, or a custom physics engine) capable of simulating realistic underwater dynamics, sensor readings with noise, and communication limitations.

• Integrate Llama 3.1 (via OpenRouter or a locally served instance) to interpret fused sensor data, assess potential threats based on contextual cues, and dynamically re-plan UUV trajectories or search patterns according to observed anomalies and mission objectives.

• Develop a BentoML service for deploying a lightweight anomaly detection model (e.g., an Autoencoder or One-Class SVM) that processes raw sensor data onboard simulated UUVs for initial threat flagging before transmitting aggregated information.

• Orchestrate communication between simulated UUVs, a central command node, and the Llama 3.1-powered decision engine using a publish-subscribe messaging system (e.g., ROS2, DDS) for real-time data flow.

• Optimize the overall system for low-latency decision-making and robustness against sporadic communication outages and sensor failures, which are critical for real-world undersea operations.

• Implement visualization tools to display UUV trajectories, sensor coverage, detected anomalies, and the AI's real-time threat assessments and generated plans.