The Gulf Spill's Lessons for Robotics

The demands on the largest underwater robotic armada ever fielded show that ROVs need better automation

4 min read
The Gulf Spill's Lessons for Robotics


Photo: Petty Officer 3rd Class Patrick Kelley/U.S. Coast Guard

Deep-Sea Command: From shipboard control rooms, ROV pilots use joysticks to steer robots through deep water and maneuver their plierlike mechanical hands.

In the weeks following the explosion of BP's Deepwater Horizon oil rig on 20 April, a dozen robots the size of moving vans descended into the Gulf of Mexico. Each tethered to a ship by a combination electrical and optical cable, the remotely operated vehicles (ROVs) formed a fleet of unprecedented size.

Deep-water-drilling companies routinely enlist ROVs to maintain and assemble equipment underwater. But in the aftermath of the explosion, BP's attempts to contain the gushing well have pushed these machines to the limits of what they were built to do.

"No one's ever seen anything like this before—that many ROVs simultaneously working on one project," says Tyler Schilling, president and CEO of Schilling Robotics, based in Davis, Calif., which manufactured four of the ROVs in the Gulf and all the robots' manipulator arms. But if predictions about the growth of deep-water drilling prove accurate, big fleets of robots will become the norm, and with that will come the need for much better automation.

Experts mostly agree that the ROVs in the Gulf have carried out their tasks with impressive success, and it is unlikely that better ROVs would have solved the crisis sooner. They have provided the hands and eyes of the entire underwater response operation. For example, when a device inside the rig's blowout preventer failed to automatically seal off the spewing drill pipe, engineers sent ROVs down to jam it into place. When that didn't work, they sent ROVs to saw off the busted pipe, position a four-story dome over the well, and later install a smaller oil-collecting cap in its place. "In those kinds of water depths, nothing happens without an ROV," Schilling says.

Sending human divers below 200 meters is risky and expensive. BP's gusher sits at 1500 meters—easily reachable by ROVs, which can work at depths as great as 7000 meters when equipped with blocks of syntactic foam. The blocks, made of epoxy and glass microspheres, compose much of the robot's bulk and keep it buoyant.

A "work-class" ROV requires a lot of power to drive its hydraulic pumps, which spin thrusters and animate manipulator arms and tools, allowing the robot to haul half a metric ton. Electricity, at as many as 3600 volts, flows from a generator on board a surface ship to the ROV through its massive tether. Unwieldy and cumbersome beasts, tethers stretch as far as 8 kilometers and weigh up to 15 metric tons, about three times the weight of the ROV itself. "Most of the energy in piloting an ROV goes into moving the cable through the water," says Craig Dawe, chief ROV pilot at the Monterey Bay Aquarium Research Institute (MBARI), in California.

Work-class robots make up less than a third of the world's ROVs, but they are the industry's fastest growing sector. Since shortly after the Arab oil embargo in 1973, the global work-class ROV fleet has grown from just three to more than 700. Texas-based Oceaneering International dominates the market with 253 ROVs and is supplying most of the robots at the spill site in the Gulf. Despite the BP disaster, analysts expect deep-water oil production worldwide to rise from 6 million to 10 million barrels a day within five years. And that will drive the total number of work-class ROVs to 1250 by 2014, according to market analysts at Douglas-Westwood, in Canterbury, England. By then work-class ROV manufacturing and services will be a US $3.2 billion business, says the firm.

Almost all such ROVs serve oil and gas companies. (The remainder maintain subsea telecom cables, aid scientific research, and mine for diamonds.) Most offshore operations need just a few robots for construction and maintenance—laying cables, operating valves, and anchoring equipment, among other tasks.


Photo: Schilling Robotics

Seafloor Workhorse: Work-class ROVs are designed to do power-intensive work hundreds of meters below the surface. Click to enlarge image

As companies expand operations with deeper wells and horizontal drilling, "facilities on the seafloor will get more and more populated [with equipment], and more and more complex operations will have to be run," says Julio Guerrero, a mechanical systems and robotics expert at Schlumberger-Doll Research Center, in Cambridge, Mass. "That is what will require development of more sophisticated technology," Guerrero says.

This includes more sophisticated robots. "ROVs will be called on to do more varied tasks and a greater proportion of them," Schilling says. They will likely work in larger numbers and in closer proximity, not unlike the congested operation unfolding around BP's blown-out well in the Gulf of Mexico.

And with so many ROVs working in such close quarters, mishaps are more likely. In early June, two ROVs collided, dislodging a tube inserted into a riser pipe. Later that month, an ROV likely nudged a valve shut on the containment cap that was siphoning oil to the surface. The cap had to be removed for a day and repaired. "There are an unbelievable number of ROVs operating down there," retired Coast Guard Admiral Thad Allen told reporters after the incident. Two setbacks in two months of work "is a pretty good record."

But some ROV experts think this record could be improved. The solution probably won't involve engineering new hardware but rather developing more sophisticated software.

ROVs make mistakes most often because their human pilots do. "Even tightening a nut with a standard combination wrench is really, really challenging for those guys," says MBARI's Dawe. "There's no tactile feedback, no depth perception, no audio feedback of what's going on down there."

To help eliminate human error, ROV manufacturers like Schilling Robotics are developing computer software to automate some of the standard things that ROVs do, like testing a rig's blowout preventer. "Automation techniques will improve not only the time that it takes to do these tasks but also the quality of the results," Schilling says. Most ROVs are already programmed to use pressure and depth gauges, compasses, and Doppler sonar to orient themselves underwater. To minimize costs, Schilling hopes to make the upgraded automation operate using only the cameras and sensors already installed on ROVs.

Automating ROVs could also refine their awareness of what surrounds them, a feature that might have been useful to the robots navigating cables and moving gear in the Gulf, says Andrew Bowen, director of the National Deep Submergence Facility at Woods Hole Oceanographic Institute, in Massachusetts. "It's an incredible ballet they're engaged in down there," he says.

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman

This article is part of our special report on AI, “The Great AI Reckoning.

"I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓ Show less