FISH FACTS: Swimming robots

Comments Comments

ELECTRONICS have changed our lives in many ways, and fishing is no exception. From the electronic fuel injection which increased the reliability and fuel economy of your outboard motor, to the satellites which give you GPS, to the sounder and sonar technology which makes finding fish a breeze. But what about the fish? No, I’m not actually talking about real fish. I’m talking about electronic fish. In a column in the recent past I reported on how US scientists had developed a robotic fish to monitor pollutants and other water quality parameters in lakes rivers and oceans. Yes, a remote controlled fish complete with onboard sensors and wireless capabilities, designed to patrol for algal blooms, oil spills and other pollutants.

The environmental scientists’s goal was to develop an artificial fish that could swim just like real fish. They did this by powering the fish with fins, making them much quieter and more natural than any other type of underwater vehicle powered by a propeller. Of course, this close replication of real fish invites the potential for unwanted interactions with predatory fish (and possibly even angler interactions on a slow day).

Perhaps predictably, these advances in electronics have continued unabated and now I report on another recently published paper in “Science Robotics” journal which described “Implicit coordination for 3D underwater collective behaviors in a fish- inspired robot swarm”.

What this actually means is that the scientists developed electronic fish (called bluebots) with a simple array of visual sensors which allows them to intercommunicate based on the sensing of blue light. Equipped in such a manner, the electronic fish were able to exhibit various complex coordinated swimming behaviours, including schooling, grouping and ungrouping, as well as search-capture maneuvers.

The ability of these simple robots to undertake these activities with minimal sensor input actually provides some insight into the behaviour of real fish. This is because it shows that many activities that fish perform in groups, such as shoaling, schooling, and predatory attacks on other fish, does not necessarily need any complex understanding, leadership or underlying co-ordination between individual fish. Instead, a simple algorithm based on visual input and adhesion to basic rules regarding minimum and maximum distances between individual fish got the job done. Of course, relying on visual cues would only work when water visibility is good, however there are electronic alternatives to generate the same information in low or no visibility situations (by using, for example, sonar) which mimic other senses possessed by real fish, such as the lateral line.

Of course, as is often the case with cutting edge technology, militaries around the world have become interested in using similar breakthroughs in electronics to develop their own co-ordinated swarms of autonomous attack drones which can seek out and destroy military vehicles, or even individual people using face recognition technology. And anyone who has witnessed one of the new drone light shows that accompany fireworks displays has already seen the same technology commercialised for nighttime entertainment. Indeed, drone light shows have been touted by their proponents to replace fireworks completely in some circumstances, such as when forest fire danger is high or when threatened native animals react adversely to firework noise.

So, if an electronic fish can be programmed to display schooling swimming behaviour, then what is stopping scientists developing an electronic fish that is programmed to eat a lure? Perhaps it might need to be a particular type of electronic lure, which the electronic fish could be programmed to detect and consume once its cast within, say, bluetooth range. Or maybe the lure only needs to be placed in the visual field of the electronic fish, but might need to be worked in a particular way in order to get a strike. But, assuming all this is technically feasible (and it almost certainly is), the next logical question is, could fishing for electronic fish be made interesting and rewarding enough to become an acceptable substitute for fishing for real fish?

Of course, electronic fish would not be edible, but then again, neither are many gamefish, whilst catch and release is very popular in many fisheries. Time and time again angler surveys have found that for many people, the actual catching and keeping of fish is secondary or tertiary to many other aspects of the angling experience. Often, it’s the other things related to the surrounding environment, the journey, or even just getting some valuable time away from work or with family and friends that are considered most important. Under those contexts, angling for electronic fish may not be such a silly concept at all.

What do you think?

The paper on schooling robot fish can be found at https://www.science.org/doi/10.1126/scirobotics.abd8668, and a video of the process can be found at https://www.youtube.com/watch?v=MRtf69ZJ41Q.

comments powered by Disqus