Managing the Transition: Matching Army Robotics Force Structure and Strategy
Lieutenant Colonel Stuart Hatfield is the Robotics Branch Chief, Dominant Maneuver Division, Office of the Deputy Chief of Staff, G-8, Department of the Army in the Pentagon, where he manages the Army’s $800 million budget for Robotics and Unmanned Ground Systems. LTC Hatfield is the Army Staff lead integrator for Unmanned Ground Systems, and he co-chairs the Joint Staff Unmanned Ground Systems Integrated Product Team to synchronize concepts, requirements, technology, and standards for remote and autonomous systems across the Department of Defense. LTC Hatfield was honored by the National Defense Industrial Association as the 2012 Ground Robotics Champion.
Interview by UTS Editor George Jagels
UTS: Please discuss your office’s purposes and how you plan on achieving your goals going forward.
LTC Hatfield: The Robotics Team within Army G-8 Force Development is responsible for the modernization strategy and managing the equipping and modernization budgets for unmanned ground systems (UGS) and robotics in the Army. We manage the research, development, test, and evaluation and procurement money that the program managers use to continue tech development; engineering, manufacturing, and development; and fielding of those systems. These funds are separate from the science and technology (S&T) budgets used by RDECOM and research labs managed by the Assistant Secretary of the Army for Acquisition, Logistics, and Technology (ASA/ALT).
There is a plan to grow the office before the next budget cycle, but right now it’s just me.
Our purpose is to provide a modernized force equipped with affordable, modular, interoperable, and increasingly autonomous UGS, enabling manned-unmanned teaming with improved protection, persistence, and endurance for the warfighter.
Our focus for new programs is to address the priorities of:
1) Protect the force at increased stand-off distances from the threat and hazards
2) Persistently monitor a changing, complex, operational environment
3) Lighten the warfighter’s physical and cognitive workloads
4) Sustain the force with increased distribution, throughput, and efficiency
5) Facilitate maneuver in wide area security and combined arms operations
6) Conduct lethal and non-lethal engagements where manned systems are limited, denied entry, or unavailable
To accomplish this, we are pursuing common chassis robots using modular mission payloads and common controllers to maximize both efficiency and effectiveness across the Army and joint services.
UTS: Which systems will be brought back from Afghanistan and reset? How is this decided?
LTC Hatfield: The process used involved a close examination of what enduring requirements the Army will have as we move towards programs of record. In the case of a small, individual transportable system, we’re working on the Common Robotic System-Individual (CRS-I, or “Chrissy”) program of record—a backpack-able system with modular mission payloads used by infantry, engineers, EOD personnel, MPs, and Special Operating Forces. This will replace the Small Unmanned Ground Vehicle (SUGV, terminated in April 2013) at half the cost and half the weight. So if that’s the long-term program of record, which of the non-standard pieces of equipment in the downrange inventory best bridge to that capability? Which systems have a residual lifecycle capability in terms of their lifecycle?
By residual capability, I mean that these systems typically have about five to 10 years of use in theater (with refresh). For example, a system purchased in 2005 and fielded and is worn out, then it has no residual capability. If it’s relatively new or can be refreshed and get another three to five years of lifecycle in order to serve as a bridging solution before CRS-I is fielded to units, then that system became a candidate for reset and recapitalization. We’re doing this for all Packbot and below systems to meet the capabilities for dismounted warfighters.
For the Man Transportable Robotic System (MTRS) Increment II [enduring capability]—which is a Talon-sized system—we have a plethora of Talons, Packbots, and others that are being used and can bridge towards this capability.
The last part of the process was a cost-benefit analysis of how much it costs to take a non-standard piece of equipment and make it a standard piece of Army equipment, which requires full material release and type classification. The cost for each type of system is anywhere between $2 and $7 million, comprising of additional testing, safety release, vendors writing operating and repair manuals, and putting repair parts in the Army supply system. So if we had only 12 of a particular robot, it was probably not cost effective to maintain that fleet and transition it into standard equipment. On the other hand, if we had a fleet of 1,200 systems that were good bridging candidates with residual lifecycle, then there was a cost-benefit potential to type classify that system and maintain it.
Of the 5,500 systems we had downrange, the Army planned on keeping 2,700 until we got the bill from ASA/ALT regarding the need to perform type classification. After that, we decided to sustain 1,400-1,500 systems in soldiers’ hands as bridging [solutions] until we get the long-term systems.
UTS: In what ways will already-purchased systems be refurbished?
LTC Hatfield: One of the largest concerns was frequency spectrum. The frequencies the Army was allowed to use in Afghanistan was not compatible with what the Federal Communications Commission (FCC) allows for use in the United States. So, at the minimum that requires a radio refresh or radio replacement on certain systems.
Another aspect is where we can, we will continue to pursue our modularity strategy with interoperability and apply our Interoperability Profiles (IOP, which define the electrical, mechanical, and logical interfaces between the modules and components of the system) towards these systems so that when we upgrade new components, we will be able to facilitate competition within the upgrades. As an example, Army EOD initiated our standardization program for EOD robots—MTRS Increment I—and came up with a procedure to bring back robots (i.e., Talon by QinetiQ) with residual lifecycle, pull the components out of it, and reinstall new interoperable and modular components competed by component rather than the whole system being competed. Small businesses can take advantage of this: Instead of going to the original equipment manufacturer for the entire system and all components, if we use that interoperability profile, a small business that specializes in arms or radios can now compete for and win the contract for a select piece of that system.
In this way we are making the system better than new and cost effectively recapitalizing the investment of the chassis that we’ve already purchased as opposed to buying all new robots. Because these systems are coming out of theater, this type of recapitalization is being done with overseas contingency operations budgets, in accordance with guidance the Army has received from Congress. This also avoids us having to transition this financial burden onto the base budget.
UTS: What is the Robotics Enhancement Program?
LTC Hatfield: We had a question from senior leaders [to the effect of]: Since robotics are such a rapidly innovating field, how does the Army continue to inform itself of what is readily available before making a major commitment (i.e., setting up a program of record to buy a thousand of the systems)? The length of time it takes stand up a program of record is from three to seven years. [In terms of innovation timelines,] robots are similar to laptops—you dispose of those every two or three years—but robots are much more expensive.
So we used the Soldier and Marine Enhancement Program mandated by Congress in 1989 as a model to set aside funds to do a “buy, try, decide” methodology. In this case, anyone can come in with a non-developmental item (commercial off-the-shelf, government off-the-shelf, or other mature technology systems) and recommend that the Army buy some small quantity of the systems to evaluate them, similar to a rapid acquisition but in very limited quantities. This allows us to stay abreast of the state of the art in industry and inform an emerging program of record or transition that capability into a program of record, while helping to cut down on the timelines for developing [that program].
The Robotics Enhancement Program (REP) is also a response to industry’s frustrations that we had hosted three robot rodeos since 2009—costing them tens of thousands of dollars to participate in—which didn’t yield contracts, programs, or a return on investment. The REP is a way to allow for a small return on investment because we purchase the systems from the company and use the program’s funds for the evaluations and safety releases on the systems (rather than the company). This also demonstrates good faith and maintains open lines of communication with industry. The REP helps to bridge the gap between the technology they have and the capabilities we are looking for … [This is important because] it can affect concepts of operations and how soldiers do their jobs.
This program will begin in 2015 and be managed by the Maneuver Center of Excellence at Fort Benning and Program Manager Force Projection under the Program Executive Office CS&CSS.
UTS: Please discuss efforts to attain commonalities within and between classes of robots.
LTC Hatfield: Commonality begins with the interoperability profiles. Between the Army’s Robotics System Joint Program Office and Navy Advanced EOD Robotics System (AEODRS) [architectures], we developed interoperability profiles to define the interfaces between the systems and between the modules on the systems. What’s happening within the module is not our concern; that’s the intellectual property of the manufacturers.
In 2010, we came together with industry to compromise on an industry-wide standard for some of these interfaces. The analogy I use is the computer mouse. There are many different types—cheap ones, expensive ones—but you know when you get it home it’s going to work as long as you have a USB interface. With robots, we have the common chassis and common interoperability interfaces. A small company that builds robotic arms used to have to bet the family farm on which large manufacturer interface they were going to partner with, whereas now with the common interface interoperability profile they can build the arm up from there and it will be able to plug and play with any system in those classes.
[In the self-transportable category,] the emerging program of record is the Squad Multipurpose Equipment Transport, which is a squad follower system designed to carry the squad load. However, it can do many more things, such as mounting a collection of engineering tools for route clearance and marking, off-load power, non-standard casualty evacuation, and network extension with larger, heavier radios. There are also options beyond that [eventually] with tele-operated fire support with a larger weapons than squads currently carry. That is not, however, imminent.
UTS: Some of the far-term projects your office plans for are decades away. Can you provide insight into how these projects are decided upon?
LTC Hatfield: We have a process called the long-range investments requirements analysis, which is a 30-year strategy where we look out for what capabilities will be needed, take input from the S&T community about when technologies will be ready, and then apply an affordability constraint [regarding future budgets and resource prioritization] to attain those capabilities for the warfighter. This year we’re looking out to 2046 to see when technologies will be developed and how they will affect the way soldiers do their job in an emerging, complex, and highly uncertain environment.
UTS: How may a rebalance of forces away from desert environments affect your buying strategy?
LTC Hatfield: The Army’s focus now is the Training and Doctrine Command’s Force 2025 and Beyond. This strategy and force structure will deal with personnel cuts and determine what future formations look like within that constraint as well as accomplish the president’s directive to shift to the Pacific.
Operations in the Pacific will involve large expanses of water, small islands, mud, triple canopy jungle, and so forth— very different from a desert environment. The Army is aware of challenges it will have with communications, the limitations with unmanned aerial vehicle (UAV) coverage, the greater need for ground reconnaissance, and dealing with personnel constraints regardless of sequestration. The question is: How can we facilitate the development of robots from tools and members of the team? That requires additional autonomy and artificial intelligence [, among other things,] to enable, not burden, the warfighter.
One example currently in common use is tele-operation. When a soldier picks up a remote control to operate a robot, that soldier is not holding his or her weapon and is out of the fight. Compare that to a handler and a military working dog. They communicate through visual and vocal signals [while the soldier can carry his or her weapon at the ready]. This is the kind of relationship we’d like to see with robots so that the soldier is in the fight and robot is enabling the soldier’s protection, persistence, and endurance.
There are already jobs soldiers will not do without their robots, such as cave and tunnel reconnaissance: Better to make contact mechanically rather than personally with the enemy. As we get increasing autonomy, the robot will understand the intent in the mission and automate as many functions as it can, which will lower the cognitive and physical workload on the soldier.
UTS: Is there an autonomy requirement?
LTC Hatfield: Some aspects will come sooner than others. Lethal autonomous systems, for example, are not a focus. We’re looking at two extremes right now. With large trucks, we think we are paralleling industry efforts, such as Cadillac’s Super Cruise technology, in getting autonomous systems into less complex environments (e.g., well-marked four-lane roads). For driver assist, optionally manned, and leader-follower capabilities, we think they are progressing very quickly. Hopefully, [such systems] will begin fielding in 2020 and definitely in support of Force 2025 and Beyond.
On the other end of the spectrum, the chief of staff of the Army has asked to see a system which a solider could pull out of his cargo pocket and release for individual reconnaissance to see what’s in the next room, around the corner, or over the hill. Here’s where micro- or nano-UAVs may come in. But piloting the “bumblebee camera” is problematic, and you can’t keep your hands on your weapon while doing so. These systems will require more autonomous capability (such as simultaneous localization and mapping).
In the middle of these extremes, we want the squad follower-type systems that keep the entire squad in the fight. The level of autonomy we want will allow the system to see the soldier, follow the soldier, and communicate in a manner similar to military working dogs.
This interview originally appeared in the Fall 2014 issue of Unmanned Tech Solutions magazine.