Coralbots will be trained to distinguish coral fragments from other objects, such as sponges and other sea creatures, as well as rocks and trash. (Source: Murray Roberts/Heriot Watt University, Edinburgh)
I also wish we'd been able to see what they look like. It's worth remembering that the by now famous U of PA's flying robot quadrotor swarm learned to build things https://www.grasp.upenn.edu/success_story/grasp_lab_drones_colbert_report as did a similar swarm in France http://www.designnews.com/author.asp?section_id=1386&doc_id=249645&image_number=2 Building things is what the coral-repairing swarm will do, so it's not hard to imaging that, assuming the robots stay waterproof, they'll be able to rebuild the reefs pretty quickly, especially with enough of them.
Fascinating, Ann. I take it they would rebuild by using broken pieces of coral -- or perhaps supporting coral that is beginning to break. This could be a big deal given that coral reefs are in bad shape all over the globe.
This is a cool concept and some neat technology, but it does not restore the animals that actually build the reefs other than to give them substrate and structure. This process will not really restore a coral reef, except to create man-made structure to support sea life, and there is already ways to do this cheaper. Yes, they (coral) need substrate to attach to and sea life needs reefs for protection, but if you want some lifeless structure to act as a nursery (much needed in the oceans), then I suggest sinking more de-commissioned ships to give some structure for sea life, certainly a lot cheaper, and proven to attract sea life and create new, large coral reefs relatively fast, and create eco-tourist traffic that boosts local economies. Sunken ships are better, because trawlers will stay away from a sunken ship, allowing the sea life to flourish (only to save their precious equipment). They certainly don't care about coral reefs, and as these robots build up the lifeless reefs, the bottom draggers will come along and continue to destroy them. Additional concern: I would be curious to know how sensitive to any existing coral that are attached to the materials and structure they are creating.
Chuck and others, the robots would piece together/transplant damaged bits of healthy and living, not dead, coral and re-cement them to the larger structure to help the entire structure regrow. Here's a description from a different project attempting to do something similar via human hands in shallow-, not deep-water, coral reefs: http://coastal.er.usgs.gov/coral-transplant/
The idea is to do this before a certain threshold is passed and massive, irreversible damage occurs. In Scottish case, it's probably better described as maintenance than repair.