Underwater Robot Design for Monitoring of Offshore Energy Structures

Ensuring the health of existing and future ocean renewable energy systems is crucial for the future of these systems. These systems, if unmonitored, suffer from biofouling, cracking, spalling, and corrosion, all caused by the harsh conditions of the ocean. Traditional methods of Structural Health Monitoring (SHM) often include human divers manually searching for the defects. This method is costly, leads to a large margin of error, and is hazardous to the diver. To mitigate the risks and costs associated, the research team developed a modified underwater Remotely Operated Vehicle (ROV). The team used an off-the-shelf ROV and modified it with an additional encloser, which was mounted to the ROV. This encloser contained an additional camera and sensor to increase the capabilities of the ROV. The current configuration of hardware integrates a 4k camera, ultrasonic distance and temperature sensor, 256 GB data storage, 2 4.8 V 2000 mAh power supplies, and a Raspberry Pi 5 microcontroller. These components are stored underneath the ROV in the hardware hull, where they take data and send it ashore to be analyzed by the pilot and the developed Convolution Neural Network (CNN). The processed data identifies any defects and uses the additional sensors to obtain consistent imagery while also allowing for future autonomy of the robot.