Great Barrier Reef’s Invasive Starfish To Be Battled By Robots

Two Australian researchers have developed an autonomous starfish-killing robot, termed Cotsbot, which will soon be tested in Great Barrier Reef in Australia to check its abilities to destroy crown-of-thorns starfish (COTS) that are considered a big threat to corals. We’ve now trained the robot using thousands of images of COTS collected on the reef and the system is proving itself incredibly robust at detecting the COTS.

The crown-of-thorns sea star (COTS), with its typical crown of venomous thorn-like spikes covering its upper surface, is one of the largest sea stars.

Scientists have a couple of theories for why the starfish population has exploded, but it seems to be a combination of fewer predators – thanks to fishing and shell collecting – as well as wastewater runoff, which increases plankton blooms that serve as food for crown-of-thorns starfish larvae.

The advanced computer vision helps the COTSbot to navigate in marine water conditions and its machine learning algorithms help it identify the COTS.

To this point the job of controlling the starfish has fallen upon human divers, who kill starfish by injecting them with a solution of cattle bile. According to Dunbabin, there are not enough human divers to remove this starfish from hotspots across the Great Barrier Reef.

The researchers tested the robot, believed to be the first of its kind in the world, in a trial in Queensland’s Moreton Bay and hope to deploy it at the Great Barrier Reef in December this year.

He said the coral polyps-eating spiny starfish are responsible for about 42 percent of the decline of the Great Barrier Reef. More trials are set to take place later this month.

The new robot could be huge when it comes to reef conservation. However, back then, it required up to 20 injections to kill a single starfish.

The COTSbot is a torpedo-shaped autonomous underwater vehicle (AUV) that weighs about 65 pounds.

Past year James Cook University (JCU) scientists came up with a one-shot injection method that had proved just as effective. It is also created to operate exclusively within a metre of the seafloor, which is extremely challenging for any robot. When it encounters an obstacle, Dunbabin said it usually tries to go up and over it to continue tracking the undersea terrain, not unlike a household Roomba vacuum cleaner. Machine learning, and his vision system handled rest of the bit and COTSbot is finally a reality. “If the robot is unsure that something is actually a COTS, it takes a photo of the object to be later verified by a human, and that human feedback is incorporated into the robot’s memory bank”.