Armed UGVs

Seen here during an earlier USMC trial, the Pratt Miller EMAV, fitted with a QinetiQ control suite, will form the basis of the RCV-L prototypes used in US Army Phase 2 testing and beyond. (Photo: USMC)

Robot warriors?

The US Army’s plan to fold robotic machines into its ground forces hinges on three experiments and a robust communications network. Shephard reports on the latest developments from the Robotic Combat Vehicle programme.

Jason Sherman

The US Army is readying a fleet of new experimental UGVs along with a series of exercises over the next few years. Their aim is to assess whether such systems add a sufficient punch to warrant the service launching new acquisition efforts to formally add Robotic Combat Vehicles (RCVs) to the force.

Commanders are bullish about the potential for UGVs equipped with sensors and weapons – combined with a robust communications network – to redefine how the army fights in the future.

Robot revolution

‘Robots have the potential to revolutionise the way we conduct ground combat operations,’ said BG Ross Coffman, director of the Next-Generation Combat Vehicles Cross-Functional Team (NGCV CFT). ‘Whether that’s giving increased fire power to a dismounted patrol, breaching an enemy fighting position or providing CBRNE reconnaissance, we envision these vehicles providing commanders with more time and space for decisions and reducing risk to soldiers.’

The army has established the RCV programme to explore how the service might fold unmanned fighting vehicles into the ground force.

The objective is to refine capability requirements through a series of experiments – both virtual and live – in hopes of launching a formal acquisition programme for light and medium variants by 2023 while also laying plans for a heavy model, with a tank-like capability, at a later date.




The army has ramped up investments in its RCV project over the next five years by 80%, from $420 million in the service’s five-year plan beginning in FY2020 to $758 million in the future years plan accompanying the FY2021 budget request.

‘I think the army is serious about the need to experiment with these types of vehicles,’ said James Tinsley, an analyst at consulting firm Avascent in Washington, DC.

By putting prototype technologies in the hands of soldiers and closely collaborating with industry, the army plans to develop CONOPS and doctrine for pairing robots with troops, or manned/unmanned teaming (MUM-T). The service expects the project to: inform new ways of fighting; assess the limitations and benefits of new technologies to provide the analytical basis for future RCV requirements documents and, potentially, the launch of a new class of combat vehicle.

Fantastic four

RCV is one of four main projects in the CFT’s portfolio which also includes: the Optionally Manned Fighting Vehicle, a project to replace the Bradley; the Mobile Protected Firepower (MPF) project, a programme to deliver a mini-tank for infantry forces; and the Armored Multi-Purpose Vehicle, an M113 replacement.

The army currently envisions a need for three RCV variants – light, medium and heavy – and has drafted a preliminary set of requirements for each.

‘We theoretically know what our requirements are, but we won’t know until we put all these systems out in the mud,’ said MAJ Cory Wallace, RCV lead on the NGCV CFT.

The RCV-Light (L) is to be outfitted largely with sensors capable of coordinating with other weapon systems to deliver decisive lethality. ‘The intent is for RCV-Light to be a small, expendable platform that can manoeuvre into a place of relative advantage, rapidly develop the situation for the commander and enable that commander to deliver decisive lethality to off-board means,’ noted Wallace.

The RCV-Medium (M) is a larger platform that is considered attritable. ‘We prefer it not to die, but if it must die, that is OK because we would rather the robot die than a soldier,’ said Wallace. ‘It is a little bit more robust; its lethality system is able to defeat medium armoured threats. So where RCV-Light tops out at troops and trucks, RCV-Medium has more firepower to deal with… threats such as armoured personnel carriers.’


🜂 The NGCV CFT’s MAJ Cory Wallace stands in front of an M113 adapted for use as an RCV surrogate with bespoke lethality and sensor packages. (Photo: US Army)


RCV-M is envisioned as a direct-fire augmentation platform with more power and more space for modular payloads. Both classes will have a common chassis in order to provide a commander with flexibility to configure the RCV to the needs of a specific mission.

An RCV-Heavy (H) is envisioned to ‘bring everything it needs to the party’, said Wallace. ‘It has the same lethality as a manned armoured vehicle. It is meant to manoeuvre in tandem with a human tank or human armoured personnel carrier and provide decisive lethality from a position of advantage.’

Dependent relationship

It should be noted that when the army thinks about RCVs, the service is not looking for weapon systems that independently compare courses of action and then select one based on immediate situational awareness and larger understanding of the world.

The RCV programme intends to take advantage of decades of army science and technology efforts on ground robotics. But the service is not looking for fully autonomous systems.


The service is not looking for fully autonomous systems.


‘They will never be fully autonomous,’ said Wallace. ‘Fully autonomous means you don’t need humans at all. There will always be a human in the loop at one point in time, specifically with giving an RCV the ability to engage a target. The robot will never be able to give itself permission to engage, to use its weapon capability and offensive means.’

However, the new systems will be permitted to defend themselves, independent of the operator. An RCV will, for instance, be able to swat at inbound RPGs with active protection systems.

‘We are looking at augmented teleoperation, which means that an [RCV] is in essence a remote-controlled vehicle,’ said Wallace. ‘But it has additional capabilities, such as very limited waypoint navigation, very limited obstacle avoidance, obstacle detection.’

The army has established what it calls an ‘RCV campaign plan’ that features a trio of major live experimentations – each preceded by a pair of virtual experiments – to refine its plans for robotic vehicles.

The plan is divided into three phases, with each increasing in complexity of both vehicle type and soldier manoeuvres while simultaneously expanding the capabilities of prototype platforms.

To this end, a handful of new technologies have been deployed to introduce large remote-controlled platforms to the ground force. The first is a highly modified Bradley fighting vehicle called the Mission Enabler Technology-Demonstrator or MET-D. These IFVs are designed to be a mother ship for soldiers controlling the unmanned combat vehicles.


🜂 A highly modified Bradley, the MET-D will act as a ‘mother ship’ for soldiers operating RCVs, with four internal control stations. (Photo: US Army)


Developed by the Ground Vehicle Systems Center, Detroit Arsenal, Michigan, the MET-D project is an advanced technology effort sponsored by Army Futures Command. The prototypes are outfitted with advanced features, including 360° situational awareness cameras, enhanced crew stations with touchscreens and a remote turret for the 25mm main gun.

These vehicles provide internal operator stations to control RCV platforms. In addition, the army intends for the MET-D to be a testbed for experimenting with new technologies, specifically pulling forward capabilities from government laboratories or industry – functional prototypes that show promise, providing a potential bridge to help accelerate tech maturation. This could involve soldier feedback to validate the need, or not, as well as recommend areas for further refinement.

Surrogate stage

For Phase 1 of the RCV project, the army has modified M113 APCs to be robotically controlled as RCV surrogates for initial experimentation.

‘Phase 1 is all about validating the manned-unmanned teaming concept,’ said Wallace. ‘The goal is to begin to develop baseline tactics, techniques and procedures the army will use in adopting robotic vehicles, to further expand upon and validate the concept of robotic warfare.’

Before the COVID-19 outbreak in early 2020, the army planned a one-month RCV experiment with soldiers at Fort Carson, Colorado, straddling March and April, with a platoon-size element of the 4th Infantry Division, equipping them with the pair of MET-Ds and four M113 RCV surrogates.

These special-purpose M113s were fitted with a lethality package that can be remotely controlled, including a Picatinny Lightweight Remote Weapon Station gimbal and an electrically driven 7.62mm weapon.

Two of the four RCVs are outfitted with advanced situational awareness tools, including Aided Target Detection and Recognition as well as a Third-Generation Long-Range Advanced Scout Surveillance System. In addition, these two will have hostile fire detection location situational awareness cameras.

The initial unmanned manoeuvre capability calls for soldiers to ‘drive’ the RCV surrogate via teleoperation with continuous control of the vehicles.

The operational focus of the Phase 1 experiment will be reconnaissance missions, with one objective being to demonstrate basic obstacle detection and avoidance at speeds of 20mph (32km/h) while on road and greater than 10mph while off road. Planned terrain are roads, trails and semi-autonomous control across open terrain. Soldiers are also expected to operate RCVs through light dust, rain, snow and fog.

Each MET-D will initially be outfitted with four RCV control stations, two for mobility operations and two for weapons control. The modified Bradley will also feature a drive-by-wire kit, an autonomous mobility kit and a laser detection and ranging kit to allow an optionally manned operation. In other words, soldiers will be able to dismount and remotely control the sensors and MET-D platform. In addition, the service plans to experiment with helmet-mounted displays for indirect driving by the crew.


🜂 Textron will supply RCV-Ms based on the Howe & Howe Ripsaw M5 tracked platform. (Photo: Textron)


Planned missions include route reconnaissance, surveying obstacles, area reconnaissance and conducting a screen. The tail end of the experiment calls for MET-Ds and RCV surrogates to demonstrate a ‘future’ scenario that includes decentralised planning and execution, closed-hatch operations with two-person crews and assessing manoeuvre with maximum extension of the RCV tether.

In addition, this last phase will explore how RCV units might deal with highly challenging modern warfare technologies and tactics, including aggressive enemy low-flying UAS, contested communications, aggressive targeting and signature management.

‘We’re trying to solve simple problems first,’ said Wallace. ‘And then spiral the lessons that we learned in the previous experiment into the next one.’

Spring training

Next, the army will begin work on Phase 2, where the goal is to scale up the size of the experimentation – set for spring 2022 – from a platoon to a company-level demonstration. This effort is expected to set the stage for thinking about wide application of RCVs. ‘We know that lessons we learn from a company we can transpose on a brigade,’ said Wallace.

The 2022 event will aim to scale up MUM-T manoeuvres while also advancing autonomous performance of robotic platforms. The 2022 experiment will feature six MET-Ds controlling a dozen RCVs.

‘We’re in the process of building out those additional MET-Ds right now, there’s some additional technologies we’re considering… And we’re constructing what that experiment looks like right now,’ said Chris Ostrowski, programme manager for NGCV experimental prototyping.

In the Phase 2 experiment, the mission will move beyond reconnaissance to both offensive and defensive taskings, including demonstration of an in-stride breach using some type of robotic capability – either mine-clearing on a small robotic platform or an armoured breaching vehicle. The event also plans for RCVs to conduct standoff chemical detection using sensors integrated on one of the robotic platforms.

Breaching and chemical detection are ‘two of the most dangerous missions that we have our soldiers do’, noted Wallace; an in-stride breach is one of the casualty-producing manoeuvres a mechanised force can execute.

Rapid prototyping

In January 2020, after asking industry the previous year to quickly propose RCV variants for Phase 2, the army selected QinetiQ North America to build four RCV-L prototype vehicles and Textron to build four RCV-M prototype vehicles.

The RCV-L platform is based on Pratt Miller Defense’s Expeditionary Modular Autonomous Vehicle (EMAV), originally designed for the Marine Corps Warfighting Laboratory. The RCV-L variant combines Pratt Miller’s EMAV chassis with QinetiQ’s open architecture UGV control systems, providing the army with what company officials say is a mature high-performance platform with extremely low schedule and performance risk.

‘The EMAV offers a unique blend of tested maturity and high performance that provides our government customer with a platform they can not only use for near-term experimentation, but also one that is ready to transition to the field,’ Jon Hastie, QinetiQ’s director for unmanned systems products, explained.


🜂 Developing the RCV-H presents a number of technical challenges. One answer could be leveraging the designs produced for the MPF programme by BAE Systems (depicted) or GDLS. (Image: BAE Systems)


Pratt Miller Defense’s director of robots, Jim Fontaine, added: ‘I have no doubt that the EMAV will exceed the expectations of the US Army experimentation team. The Marine Corps Warfighting Laboratory has been independently experimenting with EMAV platforms for the past couple of years and the results have been nothing short of phenomenal. Our primary goal is to ensure the US Army has a proven platform to conduct their experimentation on, without worrying about the capability of the underlying technology.’

For its part, Textron teamed with small tracked vehicle manufacturer Howe & Howe Technologies and FLIR Systems to offer the Army an RCV-M variant of the Ripsaw M5, which the company calls a ‘fifth-generation robotic combat vehicle platform [which] combines scalable armour, suspension and drive options to meet a variety of mission demands’.

While these two contract awards give the winners a leg up in helping the army shape RCV requirements, it is not necessarily a lock-on for future business. ‘I don’t think this is the end of the competition,’ said Tinsley. ‘We have awards for these prototypes and demonstrators to be tested.’

He noted that innovative efforts outside the service could have implications for future army plans, such as the USMC’s Rogue Fires programme which is pairing an unmanned Joint Light Tactical Vehicle with a High Mobility Artillery Rocket System. ‘Some very interesting things are happening in experimentation outside of the army that might come back and influence the army. I don’t think the downselect to QinetiQ/Pratt Miller and Textron is the end of the story,’ Tinsley said. ‘I think it is just the beginning.’

Challenging task

Phase 3, the final live experiment, is slated for spring 2024. The objective here will be to use unmanned combat vehicles to execute the most dangerous ground mission, a combined arms breach.

‘The combined arms breach tends to be the most casualty-producing effort that a mechanised force can do,’ said Wallace. ‘It’s also the most complex because it requires most synchronisation. You’re trying to synchronise a direct fire, you’re trying to synchronise engineer assets, you’re trying to synchronise friendly assets, who are providing suppressive fire and then assaulting, establishing our side security. So it’s a very risky, very complex endeavour.’

The RCV force will attempt to breach an obstacle that is defended against a prepared enemy, an opposition that has planned its indirect fires, synchronised their observations and fields of view and has had sufficient time to provide a thorough and effective defensive scheme and overwatch of an obstacle. ‘So not only do you have to fight through an obstacle, you also have to fight through an enemy that is in a prepared defence,’ said Wallace, underlining the complexity of the challenge involved in the last of the three planned experimentations.

According to the army, RCV Phase 3 prototyping will build off lessons learned in the first two experiments, with the objective to develop innovative, unmanned platforms that investigate different approaches to solving challenges. Current plans call for exploring new modular mission payloads. The goal is to award at least two contracts for the design and build of 12 new RCV platforms for the 2024 experiment.

Phase 3 prototype platforms will focus on decisive lethality mission roles and will be designed for remote operation and the integration of direct fires, missile systems and advanced sensors. Remote-control software for modular mission packages, including obscuration, EW, chemical-biological and other reconnaissance sensors, will be matured and integrated.

As part of this project, the army has issued a draft capability development document (CDD) for its RCV efforts, which includes an estimate of total unit procurement in order to provide a baseline for estimating affordability factors such as average unit cost and total life-cycle cost. Typically, such a document would not be readied until formal launch of a programme of record – called Milestone B and currently set for 2023.


🜂 Successful selection for initial testing may not necessarily lead to a production contract for the OEMs supplying the prototypes. (Photo: USMC)


‘The intent is that this draft CDD becomes a running estimate,’ said Wallace. ‘So, after every virtual experiment, after every live experiment, we pull those requirements documents off the shelf, we update them from what we learned both from the soldiers and the vehicle platform’s performance so that by the time we get to a Milestone B, those requirements have been vetted by soldiers, validated through live testing, and then go through the regular staffing process… So, we’re going to have a very mature set of requirements at Milestone B.’

At this point, the army expects either RCV-L or M to launch first in 2023. ‘Whatever is the most mature will go to the Milestone B in 2023,’ said Wallace.

Weight problems

The RCV-H is much further off, service officials believe. ‘There are still a lot of problems we have to solve with the Heavy. For example, we want it to be as survivable as a tank, but it’s got to weigh 30 tons,’ said Wallace. This is an aggressive target that represents less than half of the heft of today’s 72t Abrams tank. ‘The technology isn’t ready to move forward to develop that kind of platform with those kinds of requirements. So, in order to avoid making the same mistakes of the past with the Future Combat System [the $20 billion failed acquisition from the early 2000s], we do not want to move forward with that programme until we are absolutely sure industry can deliver.’


We want it to be as survivable as a tank, but it’s got to weigh 30 tons.


While waiting for resolution of the technical challenges necessary for an RCV-H – such as an automatic ammunition loading system for a 105mm or 120mm main gun that is as fast as a human crew – the army plans to wrestle with organisational and doctrinal matters. ‘We can get after a lot of these in the virtual experiments,’ said Wallace. ‘We don’t have to wait until the technology is here.’

While army leaders are not publicly discussing an RCV-H acquisition strategy, some in industry see potential near-term candidates in the MPF programme to provide infantry units with a light tank. In December 2018, the army selected General Dynamics Land Systems (GDLS) and BAE Systems to build a dozen MPF prototypes each, based on the UK’s Ajax chassis with the M1 Abrams turret and the M8 Armored Gun System, respectively.

‘There is a potential for Mobile Protected Firepower to be adapted into that direct-fire solution,’ Tinsley said. ‘One of the questions is whether BAE and GD, who have been downselected, have appropriate hooks for unmanned operations or robotic, tethered operations. If they do, then that could become less a competition and more of an engineering development programme of the downselected MPF.’

Network necessity

While the RCV effort works to refine technology and demonstrate the potential for robotic capabilities to improve ground forces lethality, the ultimate fate of the project is in the hands of communications innovators and experts.

‘The biggest challenge we have is with the network,’ said Wallace. ‘Honestly, because we could have the best platforms in the world, the best new helicopters, the best new artillery pieces. None of it is worth anything if it doesn’t have a network – a secure digital, data-transfer layer that is cyber resilient, that is electronic warfare-resilient, that is dynamic, that can elegantly degrade and autonomously adjust to an electronic warfare or cyber attack and not degrade performance.’

The network is particularly critical to the viability of any future RCV effort because operators are trying to control UGVs up to 3km away. Resilient communications are required in order for an operator to trust the robot is going to obey a command exactly.

‘I mean, I’m not trying to be too simplistic here, but everything else is really an engineering problem that we can solve with enough time and enough money,’ said Wallace. ‘But that network is very challenging. And there’s a lot of people working on it… Before we move forward, we need to have to make sure we have a secure digital backbone. I guarantee you, we’re not going to move forward with an RCV programme of record if the network’s not ready to support it… because if anything it is going to increase additional risk in operating unmanned ground combat systems if that bond and trust isn’t there.’