
Using commercial solutions, the Navy and the Defense Innovation Unit (DIU) have demonstrated significant improvement in retraining machine learning (ML) models used on unmanned underwater vehicles (UUVs) to detect targets of interest as conditions and threat tactics change. The Navy used the new solutions in the fall of 2023 at an Integrated Battle Problem exercise, demonstrating a 97 percent reduction—from six months to a few days—in the time needed for ML model updates. The underwater target detection algorithms were installed…