Tracking Sharks With Robots

Scientists have been tracking sharks using robots for years. But a new design allows them to do this while tracking the animal. The system was developed by biologists from Mote Marine Laboratory, and engineers from Harvey Mudd College using components that were readily available.

It has serious gripping power capable of enduring pull-off forces 340 times its own weight. It can also sense changes in objects and change its course accordingly.

Autonomous Underwater Vehicles

Autonomous underwater vehicle (AUV) are programmable robotic Shark machines that, according to the design, can drift or drive through the ocean, without human-controlled control in real-time. They are equipped with a variety of sensors to monitor water parameters, and to explore and identify ocean geological features, sea floor habitats and communities, and more.

They are typically controlled from a surface ship via Wi-Fi or an acoustic link to relay data back to the operator. The AUVS is able to collect spatial or temporal data and can be used in a larger group to cover a larger area faster than one vehicle.

Like their land counterparts, AUVs can navigate using GPS and a Global Navigation Satellite System (GNSS) to determine where they are in the world and how far they have been from where they started. This information, in conjunction with sensors for the environment that send data to the computer systems onboard, allows AUVs follow their planned trajectory without losing sight of the goal.

Once a research project is completed after which the AUV will float to the surface and be returned to the research vessel from which it was launched. A resident AUV can remain submerged for a long time and conduct regular inspections that are pre-programmed. In either case the AUV will periodically surface to signal its location via an GPS signal or acoustic beacon, which is transmitted to the surface ship.

Some AUVs communicate with their operator constantly via satellite links to the research ship. Scientists are able to continue their research on the ship while the AUV gathers data underwater. Other AUVs can communicate with their operators only at specified dates, like when they require fuel or to check the status of their sensors.

Free Think says that AUVs aren't just used to collect data from oceanography but also for the search of underwater resources, such as minerals and gas. They can also be utilized to assist in environmental disaster response and assist in search and rescue operations following tsunamis or oil spills. They can also be used to monitor volcanic activity in subsurface areas and monitor the conditions of marine life, including coral reefs and whale populations.

Curious Robots

Unlike traditional undersea robots, that are programmed to search for one specific element of the ocean floor The more curious robots are designed to explore the surroundings and adapt to changing conditions. This is important, because the underwater environment can be erratic. If the water suddenly starts to heat up it could alter the behavior of marine animals or even result in an oil spill. Curious robots are designed to swiftly and effectively detect changes in the environment.

One team of researchers is developing an innovative robotic platform that utilizes reinforcement learning to train the robot to be curious about its surroundings. The robot, which appears like a child, complete with yellow jacket and a green arm, is able to detect patterns that could signal an interesting discovery. It is also able to make decisions based on its past actions. The results of this research could be used to design an artificial intelligence that is capable of learning on its own and adapting to changes in its environment.

Researchers are also using robots to explore parts that are too dangerous for humans to dive. Woods Hole Oceanographic Institution's (WHOI), for example has a robot named WARP-AUV, which is used to investigate wrecks of ships and to locate them. This robot is able recognize reef creatures and discern jellyfish and semi-transparent fish from their dim backgrounds.

This is a feat of sheer brilliance considering that it takes years to train a human to perform this task. The brain of the WARP-AUV has been trained to recognize familiar species after thousands of images have been fed to it. The WARP-AUV is a marine detective that also sends live images of sea creatures and underwater scenery to supervisors on the surface.

Other teams are working to create robots with the same curiosity as humans. For instance, a team headed by the University of Washington's Paul G. Allen School of Computer Science & Engineering is looking for ways to train robots to be curious about their surroundings. This group is part of a three-year initiative by Honda Research Institute USA to develop curious-minded machines.

Remote Missions

There are many uncertainties with space missions that can result in mission failure. Scientists don't know what time the mission will take, how well the components of the spacecraft work and if any other forces or objects could interfere with the spacecraft's operation. The Remote Agent software is intended to ease these doubts by completing many of the complicated tasks that ground personnel would be able to perform when they were present on DS1 during the mission.

The Remote Agent software system includes an executive planner/scheduler, and model-based reasoning algorithms. The planner/scheduler produces a set of activities based on time and events that are referred to as tokens that are then delivered to the executive. The executive decides on how to transform the tokens into a series of commands that are sent directly to spacecraft.

During the experiment there will be a DS1 crew member will be present to keep track of the progress of the Remote Agent and deal with any issues outside of the scope of the test. All regional bureaus should follow Department requirements for records management and maintain all documentation used in conjunction with establishing a specific remote mission.

SharkCam by REMUS

Sharks are elusive creatures, and researchers have no idea about their activities beneath the surface of the ocean. Scientists are cutting through the blue haze using an autonomous underwater vehicle known as REMUS SharkCam. The results are incredible and terrifying.

The SharkCam Team A group of scientists from Woods Hole Oceanographic Institution took the SharkCam which is a torpedo-shaped camera and to Guadalupe Island to track and film white great sharks in their habitat. The 13 hours of video footage together with images from acoustic tags attached to sharks, reveal much about the underwater behavior of these predators.

The REMUS sharkCam, built by Hydroid in Pocasset MA, is designed to track the location of a tagged animals without disturbing their behavior or alarming them. It utilizes an multidirectional ultra-short baseline navigation device to determine the range, bearing, and depth of the shark robot vacuum parts, then it closes in at a predetermined distance and position (left right, right, above or below) to film it swimming and interacting with its environment. It communicates with scientists on the surface every 20 seconds, and can accept commands to alter its speed and depth or standoff distance.

When state shark scientist Greg Skomal, WHOI engineer Amy Kukulya, Pelagios-Kakunja shark researcher Edgar MauricioHoyos-Padilla of Mexico's Marine Conservation Society and REMUS SharkCam software creator Roger Stokey first envisioned tracking and filming great whites using the self-propelled torpedo, which they named REMUS SharkCam, they worried that it could disturb the sharks' movements and could make them flee the area they were studying. Skomal together with his colleagues, revealed in a recent paper published in the Journal of Fish Biology that the SharkCam survived despite nine bumps and biting from great whites that weighed hundreds of thousands of pounds during a week of study near the coast of Guadalupe.

The researchers concluded that the sharks' interactions with REMUS SharkCam, a robot that was recording and tracking the activity of four sharks that were tagged, as predatory behavior. They recorded 30 shark mop and vacuum robot interactions with the robot including bumps, simple approaches and, on nine occasions, aggressive bites from sharks that appeared to be targeting REMUS.