A Prototype for AUV Post-Mission Debrief Generation From Metadata
Abstract
A prototype system will be demonstrated that can automatically generate natural language reports from metadata from Autonomous Underwater Vehicles (AUVs). The system takes time-series sensor data, mission logs, together with mission plans as its input, and generates post-mission debriefs for human operators in a concise and easy-to-understand manner. The state-of-the-art systems tackling similar issues are usually designed for simulated environments or can only interpret plans. To the best of our knowledge, this is the first system that directly deals with noise-prone real data from mission logs combined with mission plans.