Artificial Intelligence and Machine Learning for Acoustic Contact Localization (AIM4ACL)

In this project, Daniel H. Wagner Associates, Inc. (DHWA) will enhance our technologies that associate and fuse data from both passive and active sensors, thereby improving situational awareness, reducing data volume, and reducing data density. We expect significantly improved operator undersea warfare (USW) search performance and confidence in human-machine teaming; and also improved algorithm run time. We fully recognize the tradeoff between automated and manual (human) tasks. The goal of this effort is to make the sailors' job easier, but not to do it for them: to automate the decision aid, not the decision. We have demonstrated the value of this capability using a realistic simulation environment (provided by Raytheon as part of the DARPA Collaborative Operations in Denied Environments (CODE) Project), in prime contractor test and evaluation (T&E) environments (provided by Lockheed Martin as part of the DARPA CODE Project), and in various real-world systems (e.g., ONR Unmanned Surface Vehicle (USV) SWARM Project and SCO Sea Mob). We have also demonstrated the ability to accurately associate and fuse real-world submarine passive line-of-bearing data from multiple arrays in our Enhanced Multi-Sensor Association Service (EMAS) work for IWS 5, and the ability to accurately associate and fuse realistic passive line-of-bearing data from multiple platforms in our Unified Operational Picture (UOP) SBIR for IWS 5. In addition, we have demonstrated our ability to use detailed environmental data for tracking (including multi-path information) in our State Estimation Tool for Undersea Systems (SETUS) SBIR for IWS 5, and our ability to use detailed information concerning the uncertainty in this environmental data for tracking in our Probabilistic Sensor Performance Prediction (PROSPER) project for ONR. DHWA has industry-leading experience in designing, developing, and fielding mathematical algorithms and software with operationally effective capabilities. We are confident that this technology can enter the fleet as part of a future APB (and eventually ACB) release. Unclassified data for this feasibility study will be generated by DHWA's internal modeling and simulation software, Detailed Autonomous Unmanned Surface Vehicle Modeling and Simulation (DAMS), a joint DARPA/ONR SBIR project.