Unsupervised Learning of Terrain Appearance for Automated Coral Reef Exploration

We describe a navigation and coverage system based on unsupervised learning driven by visual input. Our objective is to allow a robot to remain continuously moving above a terrain of interest using visual feedback to avoid leaving this region. As a particular application domain, we are interested in doing this in open water, but the approach makes few domain-specific assumptions. Specifically, our system employed an unsupervised learning technique to train a k-Nearest Neighbor classifier to distinguish between images of different terrain types through image segmentation. A simple random exploration strategy was used with this classifier to allow the robot to collect data while remaining confined above a coral reef, without the need to maintain pose estimates. We tested the technique in simulation, and a live deployment was conducted in open water. During the latter, the robot successfully navigated autonomously above a coral reef during a 20 minutes period.

1. Introduction

Underwater marine environments provides for challenging problems relevant to the robotics community. A significant constraint from these environment is the electrical conductivity of salt water, preventing the propagation of radio waves. This limits the bandwidth and range available for communication with untethered vehicles. Furthermore, GPS signals can not be used for localization. Given these constraints, the level of autonomy required to accomplish certain tasks such as surveying or mapping is significantly increased, compared to land vehicles.