Development of Simulator for Autonomous Underwater Vehicles utilizing Underwater Acoustic and Optical Sensing Emulators

摘要

This paper addresses an autonomous underwater vehicle (AUV) simulator with underwater image sonar and optical vision emulation. Recently, various underwater missions have been automated through the great improvement in underwater image sonar and optical vision technology along with utilization of artificial intelligence. Development of the simulator that emulates underwater image sonar and optical vision can support intelligent underwater missions by visualizing and simulating virtual scenarios and reproducing real missions. In order to benefit from existing software, the presented simulator is based on ROS (Robot Operating System) environment integrated with our image sonar and optical vision emulators. When an underwater virtual scenario of terrain, target objects, and AUVs with sensors is plotted using standard 3-D modeling programs, the simulator configures the scenario and displays sonar and optical images. The ultrasound and light Beams are modeled as a set of rays each, and the image sonar and optical vision are modeled as objects detecting collisions between the rays and target objects at certain positions and orientations. The sonar images generated by the simulator are compared with real images to confirm the validity of the models.