Single camera depth control in micro class ROV

Abstract

Navigation is one of the main challenges in an underwater vehicle. To measure and sustain the depth in the micro class remotely operated vehicle (ROV) robot is one of the main demands in the underwater robot competition. There are many sensors that can be used to measure the depth; one of the sensors is using a single camera sensor. In this works, camera-based depth control is developed and evaluated for micro class ROV, namely as fitoplankton SAS ROV. Fitoplankton SAS ROV is a micro ROV prototype with six thrusters. To maintain the depth position, a PID control system with a camera-based depth sensor as the input of the setpoint is used. Moreover, the method for the camera to measure the distance is using the triangle similarity method. In this paper, the experimental scenario is using the rectangular marker to measure the distance, and the value of the depth is processing in the ground control station (GCS). The GCS will send the thruster value to control the depth, which depends on the PID control system. The experiment results show an average of depth accuracy of 95.74% to the depth setpoint.