Navigation System for a Coral Reef Monitoring Robot (C-bot)

Abstract

Coral reef monitoring robots(C-Bot) are used for long-term monitoring of coral reefs and are an alternative to the diver and buddy system due to shorter mission time. To compute coral cover and georeference data, an accurate underwater position of the vehicle is necessary. Ultra-Short Baseline (USBL) systems are used for the navigation of the underwater vehicle, and an Extended Kalman Filter (EKF) is used to estimate the vehicle’s position using the USBL measurements accurately. While a single beacon and USBL transponder are sufficient for range measurements, more than one beacon or USBL transponder is required for bearing measurements. The generation of acoustic signals for communication between the beacon and the USBL system is also a very energy-expensive process. We intend to find a suitable navigation method for C-Bot. In this paper, we analyze the effect of the update rate of measurements required to estimate the vehicle’s position. We also compare the performance of a range-only EKF with an EKF that uses a range and bearing measurement model.