In 2014, the U.S. Army Tank Automotive Research, Development and
Engineering Center (TARDEC) and the U.S. Army Armaments Research,
Development and Engineering Center (ARDEC) teamed up to integrate a
remote weapon system on a robotic vehicle to see if that system
could become certified on a Scout Gunnery Table VI course, the same
course used to train and qualify ground combat vehicle crews.
The vehicle was a High Mobility Multipurpose Wheeled Vehicle
(HMMWV), and its "brain" was the TARDEC-developed Robotic Technology
Kernel. ARDEC contributed the prototype wireless system known as the
Picatinny Lightweight Remote Weapon System, which it had developed.
The command-and-control HMMWV consists of the Warfighter Machine
Interface, developed in-house at TARDEC, which controls and operates
the robot and weapon system. Collectively, this Wingman capability
allows Soldiers in a command-and-control vehicle to remotely operate
an unmanned ground vehicle weapon system.
|
|
Initial experiments have met with limited success, but the
Wingman program has ignited further investigation into weaponized
robotics and how keeping the Soldier-in-the-loop could mitigate many
of the gaps seen in today's autonomous systems.
In 2016, the
U.S. Naval Surface Warfare Center Dahlgren Division (NSWCDD) joined
the Wingman team with its target acquisition and tracking system,
the Autonomous Remote Engagement System. With the addition of the
NSWCDD, the Wingman program received three years of funding to
demonstrate the technology. The program will culminate in a military
utility assessment at an Army national training center or equivalent
between 2019 and 2020. TARDEC engineers say Wingman is the research
and development (R&D) community's first step toward weaponized
robotics.
"The Wingman technology developed today will be foundational for
tomorrow's advanced fighting vehicles," said Dr. Robert Sadowski,
TARDEC chief roboticist. "The Wingman technology will extend the
warfighters' reach and direct-fire engagement range, allowing our
Soldiers to dominate more terrain while keeping them out of harm's
way."
TARDEC is leading the Wingman development effort with
technical partners ARDEC, NSWCDD and the U.S. Army Research
Laboratory (ARL), which provides the analysis necessary to assess
the Wingman technology from a Soldier's perspective for operational
and training purposes.
Military ground elements in first
contact with the enemy often uncover obstacles, suffer the highest
casualties and become decisively engaged, limiting friendly freedom
of maneuver. Capable autonomous systems could provide a tactical
advantage for these operators. However, aggressive state and
nonstate actors are also pursuing the development of armed lethal
robotics. As the level of autonomous capability increases,
automation will spiral into weaponized systems. Unmanned systems
deployed by our adversaries could impact the advantage our current
reconnaissance forces have in the fight for information and increase
the already high mortality rates of these units.
The Wingman
technology demonstration program will investigate how to use
unmanned assets to project lethality and move effectively with a
mounted formation and engage ahead of or along with manned platforms
without increasing manpower requirements. The team believes that
unmanned assets can reduce casualties by extending the reach of the
warfighter through unmatched advanced situational awareness,
platform autonomy and targeting in a weaponized unmanned ground
vehicle (UGV).
Wingman will begin to develop the concept of
operations and tactics, techniques and procedures to integrate
weaponized, unmanned systems into the current force and increase
operational standoff.
Initiating contact with UGVs gives
commanders flexibility and maneuver space to effectively respond to
enemy threats, and eliminates some of the risks of casualty
extraction. The Wingman technology will allow friendly commanders
the ability to disperse manned systems without creating exploitable
gaps and seams in their own formation.
In 1997, a computer named Deep Blue beat world chess champion
Gary Kasparov. By 2005, two amateur chess players using three
personal computers won a chess tournament against supercomputers and
grand masters. Teaming amateurs with computers produced a
significant advantage over the computers or the grand masters.
Current autonomy technologies aren't as capable at their tasks
as Deep Blue was at its in 1997. Most have gaps in the perception
and cognition areas. The use case for lethal robotic ground systems
requires a Soldier-in-the-loop in order to pull the trigger. Wingman
seeks to combine the perception and judgment of the Soldier with the
speed, power and precision of the machine to produce an effective
unmanned ground weapon system.
Currently fielded autonomous
ground systems require a high degree of Soldier oversight and tend
to be limited to a specific mission. They often fail to meet
warfighter expectations because of limitations in the autonomy or
robustness of the integrated hardware and software systems. These
constraints make it difficult to field an effective weaponized
robotic platform. The Wingman technology demonstrator will address
some of these limitations with today's autonomous technology by
developing manned-unmanned teaming behaviors to iteratively define
and decrease the gap between autonomous vehicle control and the
required level of human interaction.
"Unlike other autonomous
systems that seek to eliminate the operators, weaponized autonomous
systems will leverage the Soldier-in-the-loop to automate operations
and enhance the Soldier's reach," said Keith Briggs, TARDEC's
technical manager of the Wingman program.
The prototype
system complies with DOD Directive 3000.09, "Autonomy in Weapon
Systems," and will be used as a surrogate to inform the development
of future unmanned weapon systems.
The Wingman Weaponized Robotic Vehicle is an M1097 HMMWV and contains three primary subsystems:
First is the TARDEC-developed Robotic Technology Kernel (RTK), the autonomy system for planning and controlling the vehicle's mobility. RTK contains driving cameras for remote operation, LIDAR sensors (light detection and ranging) for object classification, stereo cameras for terrain classification, computers for computation, radios for communication, and all the essential hardware, cables and mounts. The system can be manually driven through teleoperation or autonomously driven through waypoint navigation.
|
|
The second subsystem is lethality, which uses the Picatinny
Lightweight Remote Weapon System. That system can use an M134
Gatling-style minigun or an M240B machine gun. Wingman is currently
investigating changing the M240B for an ARDEC-developed Advanced
Remote Armament System. This will provide additional capabilities,
such as an externally powered, purpose-built weapon to improve
reliability and accuracy, the ability to load and clear the weapon
remotely and an increased stowed ammunition load without decreasing
aim or stabilization.
The Autonomous Remote Engagement System
(ARES) is the third subsystem. It provides automated engagement
capabilities to decrease target acquisition time with vision-based
automatic target detection and user-specified target selection. This
system will decrease engagement time and overcome wireless control
latency through video tracking, user assisted fire-control and
control of the weapon.
The Wingman Joint Capability
Technology Demonstration (JCTD) is currently using an M1151 HMMWV as
its command-and-control (C2) vehicle. The C2 vehicle contains the
Soldier-machine interface that the Soldier uses to remotely operate
the weaponized robotic vehicle. Five Soldiers currently man
Wingman's C2 vehicle. In front sit a driver and a vehicle commander.
In the rear seats are a wireless remote weapon system operator, the
robotic vehicle operator and a manned machine gun operator through
the hatch. The Soldier in the hatch also uses a Long Range Advanced
Scout Surveillance System to designate targets and send the
coordinates to the robotic vehicle for engagement.
The C2
vehicle contains the TARDEC--developed Warfighter Machine Interface,
which provides customized interactive displays for the vehicle
commander, robotic vehicle driver and remote weapon system operator.
These interfaces will be expanded to accept voice commands to
naturally communicate with the robot and provide real-world data on
the surrounding environment.
The Wingman program will assess the performance and feasibility of
the technology against a Scout Gunnery Table VI course, which the
Army uses to train and certify crews for Army combat vehicles. The
course also evaluates the vehicle's ability to move, shoot and
communicate. Generally, a crew and its vehicle must pass the Table
VI course--during which they engage both moving and stationary
targets--annually, before participating in live fire training or
deploying. Putting a robotic vehicle through the Table VI course
will allow the team to quantify the tactical performance of an armed
UGV and directly compare this to how manned platforms perform.
During a Table VI, the vehicle crew conducts 10 engagements on
16 targets. Target ranges vary depending on the weapon system, and
target types vary from infantry silhouettes to armored vehicle
silhouettes. To pass, the crew must obtain 700 out of 1,000 possible
points. The Wingman program plans to field the first robotic vehicle
to obtain a certification on this course.
Along with hardware and software, TARDEC, NSWCDD and ARL are standing up a modeling and simulation capability through the development of a Wingman System Integration Laboratory (SIL), which will be used to develop and verify software before conducting expensive live testing. The lab also will make it easier to conduct Soldier virtual experiments to inform and develop future capabilities and train Soldiers before they use the system in live experiments on the range. The SIL integrates the real-world vehicle software within a simulated environment for rapid prototyping, software development and early assessment of interactions between the manned vehicle team and the vehicle.
Current autonomous systems face many issues in the areas of perception, cognition, classification and communications--which prevent fielding effective unmanned weapon systems, especially in hostile environments--Wingman will address these issues by exploring new ways to use the situational awareness of the Soldier-in-the-loop to supplement these capabilities and mitigate gaps in critical areas. As the R&D community's first step toward weaponized robotics, Wingman aims to reduce casualties and increase standoff for Soldiers, especially those units in first contact.
By
Thomas B. Udvare, U.S. Army Acquisition Support Center
Provided
through
DVIDS
Copyright 2018
|
|
| ||
