The soldier of the future may more closely resemble RoboCop than G.I. Joe. The U.S. Department of Defense is looking for ways in which to bring the power of robotics and advanced sensor networks to the battlefield, but the effort is still in its infancy and is starting with a project, borne from the U.S. Army Tank Automotive Research, Development and Engineering Center (TARDEC), that may eventually replace all soldier-driven tanks with autonomous unmanned ground vehicles. Such vehicles would be monitored by satellite or other wireless communication systems, via the Internet or local networks, and would likely use sensor networks to navigate themselves through battlefields and communicate with other vehicles and soldiers.
Battlefields are hostile environments. Testing and developing autonomous vehicles, therefore, begins outside the theater of operations, at military bases. Thanks to a project known as Applied Robotics for Installations and Base Operations (ARIBO), autonomous ground vehicles are also being tested and developed at locations as disparate as corporate campuses and metro transit systems.
ARIBO was born in 2011 when Jim Overholt, then the US Army’s chief roboticist, realized that developing and using semi-autonomous vehicles on Army bases could serve to improve the military branch’s knowledge base and build its fleet of vehicles that save on labor and fuel costs through the use of electrically powered vehicles. Corey Clothier, who was then working for TARDEC as a consultant, helped Overholt design and deploy the program.
“We sat in the cafeteria and developed a scrappy plan,” Clothier recalls. They considered factors such as how to enlist military bases in a program to test unmanned vehicles, as well as how they might pay for such a program. “We wanted it to be collaborative for agencies within the federal government, and avoid duplicative efforts [among different groups].”
By late 2013, the team had identified three different partners. Fort Bragg was the first partner to join. It needed a program to reduce labor needs associated with the Warrior Transition Battalion—which rehabilitates injured service men and women at Fort Bragg’s Womack Army Medical Center before returning them to service. These soldiers require transportation between their barracks and the hospital on a daily basis. While Fort Bragg had an adequate number of vehicles, Clothier says, it was having trouble staffing the fleet with as many drivers as the Warrior Transition program needed, based on the high number of wounded soldiers.
West Point joined ARIBO in order to test semi-autonomous vehicles for shuttling personnel and visitors around campus and to and from remote parking areas. Stanford University, which has a similar interest in the vehicles, signed on as well, looking for new ways in which to transport staff members and visiting scientists around the Stanford National Accelerator Laboratory (SLAC) campus.
In December 2013, ARIBO submitted an application to join the SmartAmerica Challenge, a White House Presidential Innovation Fellow project. Its goal was to help develop cyber-physical systems (which is government-speak for “Internet of Things”) and to test specific projects and activities in a range of industries.
Participating in the SmartAmerica Challenge helped Clothier and the other ARIBO developers to identify a number of companies that manufacture autonomous vehicles.
ARIBO is sourcing test vehicles from French startup Induct Technology, which makes a fully autonomous electric vehicle known as the Navia that carries up to 10 passengers and resembles a large golf cart. The Navia is programmed to travel around a predetermined route. A series of LiDAR sensors, integrated into the body of the vehicle, will send a halt signal to the motor if an obstacle is detected in the vehicle’s path. The vehicle is designed so that passengers can input their desired location through a touchscreen inside the shuttle (Induct’s marketing manager likens the Navia to a “horizontal elevator”).
Some states’ transportation departments allow fully autonomous vehicles on public streets and highways, but for now, these cars can be operated only if a human is in the driver’s seat as a backup operator. Privately operated areas, such as military bases and college campuses, are not under the same restrictions, but ARIBO partners plan to keep a human operator inside the autonomous vehicles during the first phases of the pilots, solely for the purposes of evaluating whether the vehicles can operate accurately and safely on their own. Eventually, once the team is confident in their safe operation, they will be run without operators inside them, thereby providing the labor savings that are a major incentive to deploying the vehicles.
The ARIBO partners have developed a software platform enabling the shuttles to be controlled remotely, via an Internet link made possible through a cellular modem they will install inside the Navia vehicles, in the event that some compelling reason—such as an emergency or a breakdown of the navigation program running on the shuttle—requires that the shuttle be operated by a third party. “We have an interface that is similar to Uber, in that we have a view of all the cars, we can see which are occupied, where they are going, etc.,” Clothier explains. “The remote-control system will have some minimal control over the shuttles, but we could make it pull over, for example, or send it to the depot if it has a flat tire.”
The vehicles could operate without the remote operating software, but it “enhances what we can do in terms of fleet management,” Clothier states. “And [through the software] we can use algorithms to help make the vehicles safer and more efficient.” Plus, he says, the software will allow the operator to manage and schedule the shuttles’ charging cycles, using a vehicle-to-grid interface that will allow the operator to charge a vehicle’s batteries when electricity rates are at their lowest.
The first ARIBO pilot program, at Fort Bragg, is scheduled to begin testing the Navia during the spring, with deployment—involving a shuttle carrying passengers and a dedicated operator on board—slated to take place in the fall. “We’ll start with a couple [of Navias] and will expand the fleet over time,” Clothier says. Eventually, he notes, they will no longer staff each shuttle with a dedicated onboard operator—but one staffer will monitor the remote-control software on an ongoing basis.
For Fort Bragg, using the electric shuttles will also be part of its larger efforts to become the first Net Zero energy U.S. military base. The shuttle’s batteries can eventually be powered by solar or some other means of renewable energy.
Stanford plans to begin testing its Navia shuttles in March, while West Point intends to start testing the vehicles this summer.
ARIBO was one of a handful of projects that were featured in the SmartAmerica Challenge, and has also entered the Global City Teams Challenge (GCTC), which was launched by US Ignite (a private-public partnership formed by the White House’s Science and Technology Policy Office and the National Science Foundation) and the National Institute for Standards and Technology (NIST) to focus on deploying a range of IoT applications within the umbrella of “smart cities,” which touch on transportation, infrastructure and energy management.
As part of GCTC, ARIBO has added a number of new partners looking to deploy autonomous vehicles in metro areas. These include the city of Seattle, which wants to test the use of autonomous shuttles as part of its downtown mass-transit system; the state of Ohio, which is looking for autonomous transportation options that it would deploy in partnership with the city of Columbus; the city of Greenville, S.C.; and Temple Terrace, a city near Tampa that is partnering with the University of South Florida, Tampa’s Museum of Science and Industry and the Florida Department of Transportation to test autonomous vehicles.
Two European partners include the city of Greenwich, England, and the École Polytechnique Fédérale de Lausanne (EPFL), in Lausanne, Switzerland, which are also partnering with ARIBO for the Global City Technology Challenge. All of the parties are currently developing specific use cases and technology approaches, which will be demonstrated later this year. The GCTC is a one-year program.
“The intention behind AIRBO was to get robotic vehicles to end users,” Clothier says. “Eventually, we can use the technology and learnings from these simple applications to help advance the more complicated deployments in hostile [battlefield] environments.” Since ARIBO first launched, he has founded a ground robotics consultancy called Comet Consulting, though he is still active in the ARIBO project.