Researchers Advance Artificial Intelligence Designed to Identify Optimal Locations for Marine Energy Devices and to Help Monitor Deployed Devices

A team led by Hydronalix Inc. is developing an autonomous device, called the Sea Remote Controlled Hydrographic Explorer and Recorder (SEARCHER), to monitor environmental conditions surrounding potential wave energy deployment sites, help determine optimal placement of devices, and observe potential environmental effects from deployed wave energy converters (WECs). The team is also using artificial intelligence (AI) to train the system to recognize and mark specific targets.

SEARCHER is designed to monitor conditions around a deployed device by surveying seafloor conditions and bioindicator species. Bioindicators are living things, such as certain animal species, whose presence signals the health of an ecosystem. Scientists use bioindicators, along with other forms of monitoring, to confirm how ecosystem health may be impacted by the introduction of new stimuli (such as marine energy devices).   

The team, which includes AI experts from the University of Delaware, is using AI to train the SEARCHER system to detect and map sand dollar distributions, which could be used as a bioindicator. Previous studies indicate sand dollar orientations (sticking out of versus buried underneath the sand) and depth distribution are related to the patterns of energy from waves. Sand dollars maximize the amount of food they can collect by locating in areas of the ocean where a lot of energy from waves reaches the seafloor.  The project team hypothesizes that WECs may change the amount of energy from waves that reaches the seafloor, which would lead to changes in the depth distributions of sand dollar beds (e.g., moving to shallower areas). Measurements of sand dollar distributions before and after device installation could indicate whether an ecologically significant reduction in wave energy has occurred. 

Additionally, the SEARCHER team aims to use its sonar imaging systems to detect scour depressions in the sediment around the anchor systems of deployed devices. These scour depressions can occur after a device is installed if it disturbs sediment in some way or if currents move nearby sediments after installation. If SEARCHER can detect scour depressions and sand dollars on the seafloor, it can assist with long-term environmental monitoring at WEC locations.

In Fiscal Year 2023, the team upgraded SEARCHER’s imaging system to collect higher-quality images and tested it at Patagonia Lake in Arizona, along the Oregon coast near Yaquina Bay at the PacWave North test site, and at sand dollar beds inshore of the PacWave North test site. The datasets from tests like these are crucial to help better train SEARCHER's target detection capabilities.

Following the offshore test in Oregon, the team designed and built an extension arm and mounting system that positions the imaging sensor below sea level, which alleviates interference from rough water conditions and results in clearer images. The team also tested SEARCHER's performance in offshore conditions typical of wave energy development sites and practiced imaging. 

The Pacific Marine Energy Center at Oregon State University provided logistical and at-sea support for testing in Oregon. The testing also gave students an opportunity to take part in this research. 

As additional datasets are collected and target detection efficiency improves, SEARCHER could offer new insights into the correlation between the operation of marine energy devices and the potential effects of device deployment and energy extraction on the marine ecosystem. 

This project is funded through the Small Business Innovation Research program.