Video Obtained from NYU Tandon School of Engineering
A New Kind of RoboCop
As I passed through campus today I noticed a sign detailing the negative effects that the emerald ash borer, an invasive species to North America, has upon the ash tree. In the lives of many, invasive species are never even seen as something more than a passing thought. This method of thinking is essentially quite dangerous as it severely minimizes the environmental havoc that these species promote. Thus, the need to remove these species is critical in order to ensure the survival of native species and ecosystems. However, as many know already, invasive species control seems like an impossible issue to tackle, even more so in aquatic ecosystems. In an effort to alleviate this situation, Professor Maurizio Porfiri of NYU conducted a study that used biomimetic robots to modify the behavior of an invasive species known as the mosquitofish.
Using robots for population control may sound frightening and complex at first glance, but the general principle behind this study is actually quite simple. By using a robotic replica that mimics the appearance and behavior of a specific predator, researchers can provoke a constant stress response in a target species. This stress response could trigger a number of physiological changes, such as avoidance behaviors, which are directly linked to losses of an organism's stored energy and lower rates of reproduction. By mimicking a largemouth bass, Porfiri hoped to induce a similar kind of response in a small population of mosquitofish.
For six consecutive weeks, Porfiri and his team forced small groups of mosquitofish to encounter a robotic largemouth bass for 15 minutes. As a real bass would, the “behavior” of the robot changed with each trial and even responded in real-time to the actions of the live fish. Over the course of this experiment, the researchers tracked both the interactions that occurred between the robotic/live fish and the length and weight of the live specimen each week. After compiling weeks worth of data, Porifi found that there was a positive correlation between the degree of biomimicry in the robotic fish and the level of stress response behaviors exhibited in the living organism. Some of these behaviors included not swimming, being hesitant to explore, and unpredictable movement patterns. Similarly, the fish that were exposed to a higher degree of biomimicry also displayed the highest levels of physiological stress (length and weight loss). By having forced the mosquitofish into a constant state of stress Porfiri claimed that biomimetic robot had a lasting effect on the target species making the fish less likely to survive for long periods of time and devote energy towards future reproduction.
While this is the first study of its nature, the concrete evidence it provides has major implications when it comes to finding a sustainable solution to wildlife population control. If we were to find a way to implement this in the wild, we could essentially manipulate the behavior of a particular species at will and control certain populations that are deemed risk worthy. This would allow for maximum biodiversity to be achieved as certain invasive populations could be managed by researchers to prevent them from competing with the native species. This method of population control also abolishes the need for traditional chemical methods of organism removal that might have unwanted side effects on the ecosystem or other species. Although, while I doubt that the idea of using biomimetic robots for managing invasive species is going to be implemented in the natural world, this study is still a fresh, unique, and fascinating take on how to handle a problem as old as life itself.
Posted by James Levangie (2)
I thought this topic was actually really cool and interesting. I've been overly impressed by how much progress is being made with robots and artificial intelligence. I think that if these robots were implemented into the wild, it could make things a whole lot easier. As mentioned in your post, the team was able to collect data from just leaving a robotic bass in the presence of other fish. I wonder how much more advances and discoveries could be made if we left these robots out, which would allow for researchers to not have to be there. It could make things a lot easier if they one day are put out around the world. At the same time I do see the difficulties in trying to make sure these robots are not damaged or taken if out in the wild.
ReplyDeletePosted by Matt Cayer
I completely agree, it seems like just within the last few years AI and robotics are not only becoming more advanced, but also more widely available for use in a variety of biological fields. It will be interesting to see how these researchers take advantage of such technology, which would hopefully allow for data to be extrapolated in ways never before seen! The article I wrote about seems to be the first study of its kind, as it tracks real-time interactions between AI robotics and living organisms. While using these robotic fish for population control is a lot easier said than done, I hope this study encourages other researchers to embrace and find new uses for this quickly growing technology. If this trend continues, we could eventually see AI being seamlessly integrated into both the man-made and natural environments with our own eyes!
DeletePosted by James Levangie
This is a very interesting post. Invasive species can be a massive threat to native populations of the areas they are invading into. This is a problem that is very important to the environment, and should not be taken lightly. I was not aware of robotic applications for invasive species before this article, but it seems like it could allow for a very interesting solution to the invasive species problem. Although this is a very new science, and does not have many studies behind, the few results that it does have seem very promising. This method of combating invasive species using robotics could help save native species around the globe, and help preserve the environment.
ReplyDeletePosted by John Mariano
I agree that invasive species have a significant negative impacts on the stability of ecosystems, and I also agree that using an artificial predators to increase the physiological stress in a targeted species is effective in theory. However, I am unsure of how this same concept could be practically applied on a larger scale. Also, wouldn't this artificial predator also modify the behaviors and stress responses of non-targeted species within the same environment? Humans have mistakenly introduced the wrong predators to ecosystems before, resulting in environmental disruption and massive species loss as exemplified by the extinction of 60%-90% of the native Hawaiian snail population. This application of technology is questionable in my opinion and would have to be seriously considered before being implemented.
ReplyDeletePosted by Kayla Rosiello
At this point in time, I agree with your statement that the application of this technology is questionable at best. As seen through the devastation of the native Hawaiian snail population, there are a number of potential issues that can arise when trying to manually offset the balance of the natural ecosystem. Also, the lack of similar experiments makes me question how statistically significant the data truly is. However, as with any creative idea, I am sure that the uniqueness of this subject alone will garner a significant amount of attention and follow up experiments. If these experiments also show similar correlations then maybe biomimetic robots could be introduced into small scale controlled environments. Despite this optimistic view, I doubt that in our lifetimes we will see this technology being introduced on even a local scale simply due to the immense number of variables that can be changed in an ecosystem by a single action.
DeletePosted by James Levangie
This reminds me of the class I took here at UMASS with Todd Fuller. It was an introductory level Wildlife Conservation and Management class. Todd taught us about the different methods we have for wildlife population control, which are generally divided into direct and indirect killing methods. Some examples are direct shooting, using chemicals, manipulating egg sites, etc. I don't think that the issue is not having found a "sustainable" method of population control, because we already have many different mechanisms to do so. I think the issue at hand is really how precise we can be. In Todd's class, we learned a lot about the ways that wildlife scientists have tried, and failed to maintain a population throughout history. It doesn't come as a surprise though, because we are well aware of how complex ecosystems are and how little understanding we have on the interconnectivity of all life. I am a bit untrusting of using robots or AI to do this, but perhaps AI is the way to go since it may be able to better grasp the scope/interconnectivity of the ecosystem as a whole, especially for underwater populations where is is harder for us to see what is going on. In my opinion though, population management should be the last thing on our list to fix. Various ecosystems are inevitably going to change along with climate change and that's that.
ReplyDeletePosted by Lauren Mason