EXPERIMENTAL VALIDATION OF A REAL-TIME VISION SENSOR AND NAVIGATION SYSTEM FOR INTELLIGENT UNDERWATER VEHICLES

摘要

This paper will present recent advances in the development of a vision sensor for automatic control of underwater robots. This research is a joint effort between the Aerospace Robotics Laboratory (ARL) at Stanford University and the Monterey Bay Aquarium Research Institute (MBARI). Specifically, we have developed and verified experimentally a vision sensor which provides real-time vehicle global position updates, while simultaneously creating a mosaic, or composite image map, of the ocean floor. This novel sensor is the enabling technology within our vision-based navigation system for intelligent underwater vehicles.