The Challenges and Opportunities of Underwater Exploration: A Journey with Fish Robots

As I reflect on my research in fish biomechanics, I am reminded of the importance of exploring the underwater world. Unfortunately, our current methods of studying this realm are limited by the need for long periods of time to observe aquatic life. We often rely on scuba diving or free diving to collect data, but these methods have limitations. For instance, humans can only spend three to forty minutes underwater at a time, which severely restricts our understanding of what happens in that environment.

To overcome this limitation, we are turning to robots like the fish robot I work with. This robotic platform allows us to study fish locomotion in ways that are not possible with live animals. Unlike humans, who may not always follow our instructions, robots can be programmed to mimic specific movements or behaviors. The robot's ability to change shape independently of its motions is particularly useful, as it enables us to simulate the conditions that might occur in the wild.

The fish robot I work on is a simple yet powerful tool. It consists of five servo motors connected to an Arduino, which allows me to program the robot to swim and perform specific actions. One of my research interests is how fish control their movements, particularly when it comes to pulses. By studying these patterns, we can gain insight into how fish navigate and interact with their environment.

To create the robot's body shape, I used a 3D printer to replicate the morphology of an actual fish. The result is a flexible tail that allows the robot to mimic the movements of its aquatic counterparts. The body shell itself was also copied from the Giant Danube carp, one of the larger species of zebrafish commonly kept in aquariums.

One of the key benefits of using robots like this is their ability to perform complex maneuvers. For example, we're working on implementing "concertina locomotion," which would allow the robot to change direction and move through the water in a more efficient way. This technology has the potential to greatly expand our capabilities for underwater exploration.

Of course, there are also practical applications for robots like this. The Navy, oil rig inspectors, and dam maintenance personnel could all benefit from having robots that can navigate underwater terrain with greater ease and precision than current models. These robots might be deployed in situations where humans could not safely access the area, such as inspecting underwater pipes or monitoring marine life.

The potential for these robots to revolutionize our understanding of the ocean is vast. Imagine being able to spend more time exploring the underwater world without the limitations imposed by human physiology. With robots like the fish robot, we may soon be able to explore the depths of the ocean with greater ease and precision than ever before.

As I continue my research in fish biomechanics, I am excited about the opportunities that these robots present. From studying the movements of aquatic animals to exploring new areas of the ocean, there is no doubt that robots like this will play a vital role in expanding our knowledge of the underwater world. With each passing day, we are getting closer to unlocking the secrets of the ocean – and it's only going to get more interesting from here.

"WEBVTTKind: captionsLanguage: entechnically his name is gilbert with two l's yeah right now it doesn't even have the googly eyes on it so that's why it's not working i think that's probably it i'm stephen howe i'm a fourth year phd student in the biomimicry program and i study fish biomechanics so i'm interested in how they move and how their body movements relate to their total body so this is our um fish robot it's a robotic platform that we use to answer questions about fish locomotion that we can't with live fish because robots do what you want whereas the animal doesn't always do what you ask it to and so um the robot actually provides us an interesting set of um circumstances we can change the shape of the robot independent of the motions that the body makes so if i'm studying eels and tuna i can't tell the eel would you mind something like a tuna for me and vice versa whereas with a robot we can make this shaped like a tuna and say robot please swim like an eel so there's a skeleton inside that you can then have algorithms yeah the robot is pretty simple it's five servo motors connected to an arduino and then we use the arduino program to run the robot and the program that we've written allows us to program straight swimming but also interject turns whenever we want and this is based on some of my earlier research with fish and understanding how they control their pulses so the tether it's slightly taking it for a walk right so yeah um there's no sensors or anything in here um deciding when it should turn it's just a pre-programmed algorithm and i use the tether kind of as a leash to reset the system looks like we have a motor not behaving everything on this robot save the motors and the wires is 3d printed so that flexible tail you see we printed in our machine in the back and these body shells we copied the morphology of an actual fish so this is based on a giant danube which is the larger cousin of the zebrafish that you see in the pet shop all the time we can make this shape like anything our future research is going to involve looking at how increasing body depth changes the maneuverability so think about plate shaped fish like a discus or um like a place or a skate or well places and skates are interesting because um they've turned their heads to the side and so they look like they're um top to bottom but they do swim like a laterally compressed fish but um think more like uh tangs like dory hi i'm dory where which way i'm trying to see which mother this is and i'll just unplug it sounds like yeah no that didn't drop the amperage okay there's a few different applications mainly opportunities for underwater reconnaissance it can be applied to i mean the navy would be interested in things like this but so would um uh oil rig inspections as well as dams or bridges most often uh the robots they're using right now are shaped like refrigerators and have about six squirt guns pointing off in uh several different directions and so you can imagine jets yeah you've got a very un hydrodynamic shape that's very unstable matched with a very complicated control scheme and so you lose these things all the time and they're like five million dollars a piece so if you have a control scheme that's more robust but is still just as maneuverable you can potentially not lose things nearly as often so yeah yeah um but the other thing is like just underwater exploration in general like uh we can scuba dive and free divers can dive down to like 500 feet but they can only be down there for three to forty minutes depending on the circumstances and so because of that we know almost nothing about what's happening underwater ultimately it would be good if we can spend a whole lot more time underwater and drones like this are going to be important for being able to expand our capabilities in that realm so just a single wave we're working on more complex implementations and finally it can even perform something called concertina locomotion and so once it gets to that point it'll finish a cycle so this is what they do inside of a tunnel if it hit the tunnel wall it would be detecting thattechnically his name is gilbert with two l's yeah right now it doesn't even have the googly eyes on it so that's why it's not working i think that's probably it i'm stephen howe i'm a fourth year phd student in the biomimicry program and i study fish biomechanics so i'm interested in how they move and how their body movements relate to their total body so this is our um fish robot it's a robotic platform that we use to answer questions about fish locomotion that we can't with live fish because robots do what you want whereas the animal doesn't always do what you ask it to and so um the robot actually provides us an interesting set of um circumstances we can change the shape of the robot independent of the motions that the body makes so if i'm studying eels and tuna i can't tell the eel would you mind something like a tuna for me and vice versa whereas with a robot we can make this shaped like a tuna and say robot please swim like an eel so there's a skeleton inside that you can then have algorithms yeah the robot is pretty simple it's five servo motors connected to an arduino and then we use the arduino program to run the robot and the program that we've written allows us to program straight swimming but also interject turns whenever we want and this is based on some of my earlier research with fish and understanding how they control their pulses so the tether it's slightly taking it for a walk right so yeah um there's no sensors or anything in here um deciding when it should turn it's just a pre-programmed algorithm and i use the tether kind of as a leash to reset the system looks like we have a motor not behaving everything on this robot save the motors and the wires is 3d printed so that flexible tail you see we printed in our machine in the back and these body shells we copied the morphology of an actual fish so this is based on a giant danube which is the larger cousin of the zebrafish that you see in the pet shop all the time we can make this shape like anything our future research is going to involve looking at how increasing body depth changes the maneuverability so think about plate shaped fish like a discus or um like a place or a skate or well places and skates are interesting because um they've turned their heads to the side and so they look like they're um top to bottom but they do swim like a laterally compressed fish but um think more like uh tangs like dory hi i'm dory where which way i'm trying to see which mother this is and i'll just unplug it sounds like yeah no that didn't drop the amperage okay there's a few different applications mainly opportunities for underwater reconnaissance it can be applied to i mean the navy would be interested in things like this but so would um uh oil rig inspections as well as dams or bridges most often uh the robots they're using right now are shaped like refrigerators and have about six squirt guns pointing off in uh several different directions and so you can imagine jets yeah you've got a very un hydrodynamic shape that's very unstable matched with a very complicated control scheme and so you lose these things all the time and they're like five million dollars a piece so if you have a control scheme that's more robust but is still just as maneuverable you can potentially not lose things nearly as often so yeah yeah um but the other thing is like just underwater exploration in general like uh we can scuba dive and free divers can dive down to like 500 feet but they can only be down there for three to forty minutes depending on the circumstances and so because of that we know almost nothing about what's happening underwater ultimately it would be good if we can spend a whole lot more time underwater and drones like this are going to be important for being able to expand our capabilities in that realm so just a single wave we're working on more complex implementations and finally it can even perform something called concertina locomotion and so once it gets to that point it'll finish a cycle so this is what they do inside of a tunnel if it hit the tunnel wall it would be detecting thattechnically his name is gilbert with two l's yeah right now it doesn't even have the googly eyes on it so that's why it's not working i think that's probably it i'm stephen howe i'm a fourth year phd student in the biomimicry program and i study fish biomechanics so i'm interested in how they move and how their body movements relate to their total body so this is our um fish robot it's a robotic platform that we use to answer questions about fish locomotion that we can't with live fish because robots do what you want whereas the animal doesn't always do what you ask it to and so um the robot actually provides us an interesting set of um circumstances we can change the shape of the robot independent of the motions that the body makes so if i'm studying eels and tuna i can't tell the eel would you mind something like a tuna for me and vice versa whereas with a robot we can make this shaped like a tuna and say robot please swim like an eel so there's a skeleton inside that you can then have algorithms yeah the robot is pretty simple it's five servo motors connected to an arduino and then we use the arduino program to run the robot and the program that we've written allows us to program straight swimming but also interject turns whenever we want and this is based on some of my earlier research with fish and understanding how they control their pulses so the tether it's slightly taking it for a walk right so yeah um there's no sensors or anything in here um deciding when it should turn it's just a pre-programmed algorithm and i use the tether kind of as a leash to reset the system looks like we have a motor not behaving everything on this robot save the motors and the wires is 3d printed so that flexible tail you see we printed in our machine in the back and these body shells we copied the morphology of an actual fish so this is based on a giant danube which is the larger cousin of the zebrafish that you see in the pet shop all the time we can make this shape like anything our future research is going to involve looking at how increasing body depth changes the maneuverability so think about plate shaped fish like a discus or um like a place or a skate or well places and skates are interesting because um they've turned their heads to the side and so they look like they're um top to bottom but they do swim like a laterally compressed fish but um think more like uh tangs like dory hi i'm dory where which way i'm trying to see which mother this is and i'll just unplug it sounds like yeah no that didn't drop the amperage okay there's a few different applications mainly opportunities for underwater reconnaissance it can be applied to i mean the navy would be interested in things like this but so would um uh oil rig inspections as well as dams or bridges most often uh the robots they're using right now are shaped like refrigerators and have about six squirt guns pointing off in uh several different directions and so you can imagine jets yeah you've got a very un hydrodynamic shape that's very unstable matched with a very complicated control scheme and so you lose these things all the time and they're like five million dollars a piece so if you have a control scheme that's more robust but is still just as maneuverable you can potentially not lose things nearly as often so yeah yeah um but the other thing is like just underwater exploration in general like uh we can scuba dive and free divers can dive down to like 500 feet but they can only be down there for three to forty minutes depending on the circumstances and so because of that we know almost nothing about what's happening underwater ultimately it would be good if we can spend a whole lot more time underwater and drones like this are going to be important for being able to expand our capabilities in that realm so just a single wave we're working on more complex implementations and finally it can even perform something called concertina locomotion and so once it gets to that point it'll finish a cycle so this is what they do inside of a tunnel if it hit the tunnel wall it would be detecting that\n"