New Robots Planned for Bomb Disposal Teams
But as the Navy develops the next-generation robots, technology experts worry that the effort is too reminiscent of the Army’s failed Future Combat Systems program — a multi-billion dollar project intended to modernize the ground force’s vehicles and develop robots and sensors that would all be networked. Plagued by cost overruns, schedule delays and technical difficulties, the program was canceled in 2009.
Robot developers are warning that a similar fate could be in store for the program because of the way it is structured. Its “breakout” approach of developing the new EOD robots by component and then integrating the resulting technology has a poor history in military programs. Not only does it put an already fragile industrial base at risk, but it also could cost more in time, money and effort than anticipated, they said.
“We think it’s highly risky and we don’t understand at all why they’re mimicking the FCS fiscal disaster,” said Bob Quinn, vice president of unmanned systems at QinetiQ North America, which manufactures the Talon robot.
The advanced EOD robotic system comprises a family of robots that will replace the Defense Department’s current inventory, which is scheduled to begin retiring in 2017, said Master Gunnery Sgt. Greg Carroll of the Marine Corps detachment at the Naval EOD Technology Division. The division manages the joint service EOD program and tests and approves equipment for EOD teams.
“We are looking for a giant leap in capability,” he said. Program managers are seeking robotic systems with advanced, interchangeable features that are not currently available in the commercial market.
The family of robots is broken into three increments. Increment one, a backpackable robot, will be fielded beginning in 2014. It will replace the iRobot 310 small unmanned ground vehicle. Increment two, the manual transportable robot intended to replace the iRobot MK 1 PackBot and QinetiQ/Foster-Miller MK 2 Talon, will start deliveries in 2017. Increment three, the largest robot, will replace the 700-pound MK 3 remote ordnance neutralization system made by Remotec, a subsidiary of Northrop Grumman Corp.
“Our plan is to have three different chassis and have all these interchangeable parts and pieces between the three robots,” Carroll said.
Technicians complain that while the PackBot and Talon systems are useful products, nothing is interchangeable. Moreover, they are tied to disparate operator control units.
The advanced system will comply with standards so that a camera or sensor on the largest robot could be taken off and placed onto the small backpackable system. A common operator control unit will direct all three robots, separately or simultaneously.
Unlike previous robotic development programs that homed in on a complete system, the government is opting to build the family of robots by capability.
“You might have 10 different vendors who produce a specific component and then it will be integrated together here at the technology division and tested to make sure everything is compatible,” said Carroll.
Johns Hopkins University, for example, is developing autonomous operation and haptic feedback to improve how EOD technicians interface with the manipulator arm. The goal is to make the gripper function more realistically and as similar to the human hand as possible.
“It’ll have two fingers and a thumb, basically,” said Carroll. “We want some [haptic] feedback so we know how much pressure to apply for picking up intricate pieces of an IED,” he added.
The technology division also wants to develop autonomous grasping so that EOD technicians do not have to stop a mission to switch out tools. “We want the capability to have a tool tray with different attachments — maybe a drill, a rake, a shovel — and by a touch of a button on the operator control unit, the arm swings around, picks up the selected tool and attaches it to end of gripper and does what it has to do,” said Carroll.
By splitting up development by capability, the intent is to lower the end unit cost. Planners estimate that the increment 1 backpackable robot will cost about $150,000 per unit. The hope is that the common operator control unit concept for directing all three robots will provide savings in the long run. An EOD section could deploy with five robots but only require three operator control units. “There could be some cost savings there,” Carroll said.
In theory, the acquisition plan suggests that there could be cost benefits down the road. But up front, the program is high risk, said Quinn. Buying components individually and then having a systems integrator pull them all together is an arcane procurement model — one that proved to be disastrous with FCS, he said.
“We thought the FCS system was not viewed by Congress as an effective, efficient process, but that’s what the Navy seems to be repeating, with the systems integrator getting all the money,” said Quinn.
No one is really responsible for the whole system other than the prime systems integrator under the current acquisition scheme, he pointed out. There are many technical challenges that could turn into a gold mine for the prime.
“Under this approach, all the individual components are separate components built by different manufacturers under different bids. They simply get assembled together,” he said. “If robots assembled as easily as they said, there would be a hell of a lot more in the market at lower prices. It’s not easy,” he pointed out.
The relationships between components are complex, and they are given serious consideration by robotic system developers, Quinn said. When a manufacturer purchases components for a robot, there are trade-offs that engineers have to consider in the design of the system to meet size, weight and power restrictions, among other requirements.
“What you do with the manipulator can’t be divorced from what you’re doing with the pan-tilt-zoom camera,” Quinn explained. Where engineers place the cameras and how much mass they have both impact the mounting brackets on the vehicle and the ultimate specification and placement of other payloads, such as the manipulator.
The advanced EOD robotic system acquisition model makes those engineering trades difficult, said Joe Dyer, chief operating officer at iRobot Corp. “What worries me is the inability of such an approach to make the necessary engineering trades across the full system … the trades necessary to design systems that are low weight, low cost and require less power,” he said.
There are other industrial base concerns about the Navy’s acquisition approach. “It stands in stark contrast to the way South Korea, Singapore and Israel are fostering an important disruptive technology,” said Dyer. “Those competitors on the international scene are very, very focused on industry building, and there’s some risk that [advanced EOD robotic system] pulls the other way.” The U.S. ground robotics industry already is small and fragile, he emphasized.
Timing the supply chain correctly to meet the demanding delivery schedule for the integrated systems could be a challenge. Some of the electronic components for robots have 34-week lead times, Quinn said. That is normally not a problem when companies know they have a good chance of winning a contract. “Industry will take educated risks and procure ahead, so you don’t have to wait 30 weeks for delivery,” he said.
However, in this program, vendors have to continue competing — and winning — in order to build particular components for the robots. That could bungle the supply chain process.
“A lot of companies trying to do this can’t take the risk” to procure advance materials that would only be used if they garnered a contract, he said. Because of the uncertainty, the process then becomes a “build-order business without any risk-taking” on the part of the subcontractors, he said. That in turn could potentially put the whole program in jeopardy.
“We are doubtful that it will end up being successful because of the technical risks and the financial risks,” Quinn said. Dyer, too, voiced skepticism about the success of the program.
The worst repercussion may come in the form of schedule delays. Troops anticipate that they will need increased numbers of robotic systems to contend with the evolving threat of improvised explosive devices. Any unnecessary waiting time puts them at greater risk, Quinn said.
The supply of current generation robots is limited. The Marine Corps currently allocates two Talon robots and one PackBot to each EOD section, which is composed of an EOD officer and eight enlisted marines. All of the iRobot 310 backpackable robots are in theater or at home bases for training.
When the advanced EOD robotic system is fielded, the sections will each receive two of the small backpackable robots, two of the medium manual transportable robots and one large robot.
“There is an increase in quantity and increase in capability,” said Carroll.
For now, technicians are making do with current systems and expanding their roles on the battlefield. The use of EOD robotics is not limited exclusively to IED-specific work “because at that point, they’re like a one-trick pony — just a big piece of gear you’re carrying around that you may not need,” said Master Gunnery Sgt. Christopher Behn, the EOD chief at Marine Corps Special Operations Command. “If they can employ them in other ways that are of value to the unit, then they just become a more valuable asset,” he said.
Operators have employed the robots while units are conducting raids or other operations.
“If you wanted to take a look at a compound and you didn’t want to risk sending someone in, you could send one of these robots,” said Behn. “You won’t surprise anybody or sneak up on anyone. But it gives them the ability to stand off without putting one of their own men at risk.”
A spokesperson for QinetiQ North America said that the comments by Bob Quinn that appeared in the article represent his personal views and do not speak for the entire company.
Topics: Bomb and Warhead