Unmanned Vehicles: Seeing Around The Corner And Over The Horizon

By A. Brent Strong, Deryl Snyder, and Troy Takach

The Scenario
Imagine you are a soldier in the midst of a battle. Your squad has advanced into an enemy town with the object of capturing a large cache of weapons which you believe are in a building that is two blocks ahead. Right now, however, you are under heavy rifle fire and you suspect that the enemy soldiers are hiding in the houses between you and the arms cache. They are probably firing from the windows. You also hear machine gun fire but that seems to be a block or two away. You have, in the past, heard the sounds of tanks, but you don’t know where they are. Right now you are lying on the ground behind a wall. You seem to be safe for the moment but obviously will not be successful in capturing the arms cache unless you move forward, but that seems foolhardy. Even poking your head up to see what is in front of you seems to be a tremendous risk. You know that during previous wars, like World War II, soldiers in your situation would probably ask a companion to cover them and they would jump up and run forward. (At least that is what the movies said they did.) They seemed to either get killed or win a medal of honor. Right now, however, you are not sure if that is the wisest course of action.

Just at this moment of indecision and peril, you remember that your company has been given a new tool. It seemed a bit “James Bond-ish” at the time, but now it seems worth a try. You therefore reach in your pocket and take out a small box containing a miniature, hand-sized airplane. It looks like something that you might have flown as a science project in high school. You flip the on switch, the propeller begins and you let it go up into the air above your head. You now look at the carrying box and see that it has a screen on which the street in front of you is pictured from the viewpoint of the little airplane. The box also has controls that allow you to move the plane around, just like playing a computer game. You send the plane forward and see where the enemy soldiers are hiding. You fly the plane to the next block and see that a machine gun guards the street, thus telling you that trying to move up that street is foolhardy. You then try the street on the other side and see a tank waiting to block any advance there. Hence, you must move forward, but with the knowledge of exactly where the enemy lies, that seems less formidable a task.

You use a radio to call an unmanned golf-cart sized robot vehicle that is carrying a few small tactical missiles. Upon your command, the robot moves, fully guiding itself around obstacles, into position about 100 feet behind you. Then, you maneuver the unmanned plane directly over the position of the nearest enemy soldiers hiding in front of you. Taking the GPS coordinates, you then relay those coordinates to the control unit of the robot and instruct it to fire a missile. The enemy position is destroyed, thus allowing you to advance safely. With the use of the plane to scan around corners and over the buildings ahead, and the robot to fire beyond your immediate horizon of vision, you successfully capture the arms cache.

Is this scenario taken from Mission Impossible VII? It could be! It is much closer to reality than you might imagine. The world of unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) is not just futuristic; it is already here in surprisingly sophisticated applications. And, of course, the potential is truly astounding!

Unmanned Air Vehicles (UAV)
UAVs are any aircraft that operate without an on-board pilot and typically fly with some level of autonomy. Historically, UAVs have their roots in military applications and current applications are still mostly military. Civil activities are limited but have great potential, as we will see. The U.S. military’s involvement with UAVs started in the early 1900s, and UAVs have had active roles in all major conflicts since the Vietnam War. Recently combat in Kosovo, Afghanistan, and Iraq has proven that modern UAVs are vitally important, especially in urban environments.

Current UAVs range in size from full scale aircraft down to systems that fit in the palm of your hand. These aircraft are loosely classified into three categories based on size: full scale (wing spans in excess of 50 ft), tactical (wing spans around 25 ft), and mini/micro (wing spans below 10 ft). Full scale “endurance” UAVs such as the propeller driven Predator and the jet powered Global Hawk have nearly become household names in recent years. The Predator has been used extensively in the U.S. War Against Terror (Afghanistan and Iraq), primarily for reconnaissance but also for limited attack missions. One crucial advantage of these aircraft is their ability to persistently collect data. Both the Predator and the Global Hawk can stay airborne for a day or more. To illustrate the importance of persistence, it is interesting to note that during the initial phases of the current Iraq conflict, the single Global Hawk that was deployed flew only 3 percent of the aircraft imaging-collection sorties, yet collected data on 55 percent of all air-defense related time-sensitive targets.

Predator is a light-weight, high endurance system having a wingspan of just under 50 ft, and carrying a payload of 450 pounds for up to 40 hours. It provides surveillance imagery from synthetic aperture radar, video cameras, and infrared cameras that can be sent in real-time not only to the operational commander, but directly to the front-line soldier. These UAVs have been operational since 1995, and in total have flown over 100,000 hours. The UAV ground control station is built into a single 30-foot trailer, and houses a pilot and three sensor operators. Although the ground control station must be located at the place of take-off and landing, the UAV can be monitored and controlled from anywhere in the world via satellite links. A weaponized version of Predator exists, which carries AGM-114 Hellfire missiles, and has been used for several strikes initiated by both the military and the CIA. In December 2002, for the first time in history, a manned and an unmanned aircraft engaged in combat when an Iraqi MiG-25 shot down a Predator after the Predator fired at it.

Global Hawk has a wing span of over 130 ft, and is capable of carrying a 3000-pound payload. The sensor suite contains a synthetic aperture radar, electro-optical camera, and infrared camera. Continuous flight time is 36 hours. The system is designed to perform most routine flight operations autonomously, thus allowing the operators to concentrate on mission execution while simply monitoring the aircraft performance. Communication with the ground crew can occur through satellite links, meaning the system can be remotely operated from anywhere in the world. The first developmental Global Hawk flew in 1998. Due to the terrorist attacks of September 11, several of these developmental systems were rushed into service in Afghanistan and Iraq. In total, Global Hawk flew only 15 missions during Operation Iraqi Freedom, yet provided over 4,800 images in near real-time. The system located at least 13 surface-to-air missile batteries, 50 SAM launchers, 300 canisters and 70 missile transporters; it also imaged 300 tanks. During these missions, the UAV and its sensors were operated remotely from Beale Air Force Base in California, thus reducing the logistical requirements in the field by more than 50 percent.

The next phase for this class of UAV is a weaponized system that can perform surveillance, suppression of enemy air defenses, and precision strike missions, all without a human in the loop.

Tactical UAVs include Pioneer and Shadow, with wingspans of approximately 15 feet. Pioneer gained fame during the first Iraqi conflict, where, after enemy troops realized that the presence of this UAV meant imminent 2,000-pound naval gunfire rounds would soon land on their position, they surrendered to the unmanned aircraft. Flight times are on the order of 5-6 hours, and payload capabilities are around 70 pounds. Payloads typically consist of a gimbaled electro-optical (EO) and infra-red (IR) camera and line-of-sight data links to transmit video to the ground user in real time. The all-composite manufactured Shadow UAV has been extensively used by the U.S. Army in the Afghanistan and Iraq conflicts, with the fleet totaling over 75,000 flight hours as of early 2006. These are meant as brigade-level UAVs to provide support to the ground maneuver commanders, with the advantage of providing battlefield awareness at a cost approximately 1/100th that of a manned platform, and with significantly reduced risk to human life.

The mini/micro air vehicle (MAV) class is an area of intense growth and interest. Loosely, “mini” refers to wing spans from approximately 10 feet down to 6 inches, while “micro” refers to wingspans below 6 inches. These aircraft are designed to fill the need of short-range limited-duration “over the hill” reconnaissance and situational awareness. Likely the most widely employed aircraft of this class is the Pointer, with a wing span of approximately 8 feet and a flight time of 90 minutes. This UAV carries EO or IR cameras and a short range (approximately 10 kilometer) line-of-site communications package. The Pointer system, including ground station, fits in several hand-carried cases and can be transported and operated by two users. Similar, more recent systems, such as Raven and DragonEye are similar in operation and capability, but with wing spans on the order of 4 feet. The requirements for this class of aircraft include a wide range of operational environments, simple user operation, all-weather operation, automatic collision avoidance, and low cost. An eventual military goal for MAVs is to provide personal situational awareness for every foot soldier. As part of their normal gear, each soldier would carry one or more air vehicles and a wrist-mounted control station and video screen not much larger than a common wrist-watch. These MAVs would perform under supervised autonomy, where the user would provide only high-level commands such as takeoff/land, waypoint path following, loiter, etc. For combat in urban or forested environments, one could imagine the soldier having his/her own personal eye-in-the-sky automatically loitering over their current position and providing real-time video of the tops of nearby buildings, or the next clearing. For this to occur, the airframe and ground station must obviously become even smaller. BlackWidow is probably the most well-known micro-sized vehicle, with a wing span of approximately 6 inches. Other aircraft, such as the one developed at the University of Florida, are even smaller. Aircraft of this size have demonstrated the ability to locate and image in real time a 1 meter target from 500 meters away.

UAV materials and construction are as varied as the aerodynamic designs. It is apparent, however, that composites will continue to be the material of choice for UAVs of all sizes. In addition to being lightweight, composites offer quick build time, fewer fasteners, and resistance to corrosion (from salt water, for instance). UAVs of the mini class are commonly constructed of metal and/or plastic skeletons with Kevlar skins. Larger UAVs, on the other hand, employ primarily fiberglass and carbon fiber composite materials.

The future of the military market appears solid. The U.S. UAV capabilities have expanded from a single “eye-in-the-sky” in the early 1990s to persistent intelligence surveillance and reconnaissance today. The excitement in the military market is not unique to the U.S. According to the latest “UAV Roundup” by the American Society of Aeronautics and Astronautics, as of 2005 over 500 UAV platforms were on record, and over half of the nations in the world had some type of UAV in their arsenal. Funding by the DOD for UAV development and procurement has quite literally exploded in the past 5 years, from an annual budget of $363M in 2001 to an expected $3.5B in FY09. The bulk of this funding has been earmarked for Predator, Global Hawk, and Joint Unmanned Combat Air Systems (J-UCAS), formerly known as Unmanned Combat Air Vehicle (UCAV), but smaller systems stand to benefit as well from this significant increase in spending.

One very active participant in UAV development is Astraea (Autonomous Systems Technology Related Airborne Evaluation & Assessment). Astraea is a national program in the United Kingdom (headquartered in Wales) with a mission to develop UAV systems. A wide variety of companies are partners in this venture including such well known aerospace firms as BAE, EADS, and Rolls Royce. Recent major investments in Astraea indicate the strong emphasis being placed on UAVs in the United Kingdom for both the military and commercial markets.

Potential commercial applications are widely varying, including border patrol, police surveillance, search and rescue, forest fire and wildlife monitoring, detection of fish schools, large facility security, mail and package delivery, and even recreation. Aerial surveillance currently performed by manned helicopters and aircraft represents an application where unmanned aerial vehicles could have a large impact. For instance, law enforcement in U.S. cities with over 250,000 people cumulatively use nearly 500,000 flight hours each year of piloted aerial surveillance. At a cost of $500 per hour, this represents $244M in spending each year. Statistics are very similar for traffic/news applications.

Endurance UAVs equipped with electro-optical and infra-red cameras could autonomously fly along the border and detect illegal crossings at remote locations. Large industrial facilities such as oil refineries might use UAVs to perform aerial inspections of the sites, including visual, audio, or chemical monitoring (when mounted with a suitable chemical sensor system). Such chemical systems open the possibilities of using UAVs to enter highly hazardous areas such as damaged nuclear power plants, chemical leak zones, war areas, and other places where the environment must be monitored before actual entry can be made. It is even possible that UAVs with IR or other sensors could look for survivors of disasters like avalanches, floods, and earthquakes in areas that are too remote or hazardous for manned surveillance. Super sensitive chemical detectors could detect the existence and nature of terrorist bombs and, perhaps, even effect the removal of such devices.

Current technology even allows for an auto piloted toy airplane with a wingspan of 12 inches and an on-board video camera to enter the market below the $100 price point. (So, an unskilled 10 year old boy could send his toy plane down the street to “survey the wildlife” at the neighbor’s backyard pool!)

The primary hold-up for this market are as-of-yet unresolved issues of operating unmanned vehicles in the commercial airspace. Several concerns have delayed the process, including autonomous collision avoidance and the possibility of malicious overriding of the autopilot system. The urgency of appropriate regulations is well-recognized, however. In 2004, the Defense Science Board Task Force on UAVs issued a report calling on all U.S. government agencies to work toward making UAVs an official part of military and civil aviation. “The DOD has an urgent need to allow UAVs unencumbered access to the National Airspace System (NAS) outside of restricted areas……here in the Unites States and around the world.” As soon as these regulations are established, the commercial UAV market is destined to explode.

The “brains” of these UAVs is the autopilot, which is becoming ever-more sophisticated and capable, and vary as widely as the applications for computers themselves (especially artificial intelligence applications). As UAV systems push to smaller and smaller form-factors, the autopilots themselves are becoming smaller and smaller. For example, the Kestrel™ autopilot developed at Brigham Young University and licensed to Procerus Technologies® shown in the accompanying figure weighs in at less than 17 grams, yet can control an aircraft through a fully autonomous mission. (If you are brave, you might enjoy the novel Prey by Michael Crighton, Harper Collins Publishers, 2002, in which UAVs with artificial intelligence evolve into an independently operating swarm that menaces humans.)

Unmanned Ground Vehicles
Imagine the following situation: You are driving on the highway and pass a moving vehicle that possesses no human driver. You are shocked and move a lane away just so that you are safe from this strange vehicle. You are, however, captivated by the concept of a driverless car and you want to observe it. You therefore slow the speed of your car to match the other vehicle. After cruising together for some time, you discover that the unmanned vehicle performs precisely as would a vehicle driven by a human driver. Moreover, you discover that the unmanned vehicle is not being guided by a remote operator (there are no radio transmissions to or from the vehicle). The vehicle is, truly, self-driven.

Is that scenario frightening to you? Does it seem impossible? Hardly. A similar situation actually occurred in May 2006 as engineers from the Utah-based company DesignJug® tested—in remotely piloted mode—one of its unmanned vehicles in the western Utah desert. DesignJug® belongs to a small but growing group of companies that has recognized the vast potential of the unmanned land vehicle market, which is on the cusp of redefining the way humans operate on the Earth’s surface.

The Grand Challenge
Spurred on in part by the success of unmanned air and underwater systems, the United States Congress, in conjunction with the United States Department of Defense (DoD), has mandated that one-third of all military land vehicles be autonomous (unmanned) by 2015, and two-thirds by 2025. In response to this mandate, the Defense Advanced Research Projects Agency (DARPA)—the central research and development organization for the DoD—created the DARPA Grand Challenge as a field test intended to accelerate research and development in autonomous ground vehicles. DARPA’s purpose in holding the Grand Challenge was to bring together individuals and organizations from industry, the research and development community, government, the armed services, academia, students, backyard inventors and automotive enthusiasts in the pursuit of a technological challenge. The end goal would be to help save American lives on the battlefield.

The first Grand Challenge was in held in 2004. The goal was to conduct a field test of 15 autonomous ground vehicles that would run from Barstow, California to Primm, Nevada. The winner would receive $1 million. However, all the vehicles foundered in the desert. No vehicle traveled more than eight miles.

Undeterred, DARPA sponsored another race—the Grand Challenge 2005. More than 190 teams entered the competition, with 23 competing in the final event. Five teams finished the 132-mile course over desert terrain. All of the vehicles were totally autonomous—that is, no radio guiding was allowed. The vehicles had to follow an unknown course without human assistance.

Due to this incredible advancement in technology and vehicle capability, DARPA announced in May 2006 that it would hold the Urban Challenge 2007, which will feature autonomous ground vehicles executing simulated military supply missions safely and effectively in a mock urban area. (Remember the scenario given at the beginning of this article.)

“Grand Challenge 2005 proved that autonomous ground vehicles can travel significant distances and reach their destination, just as you or I would drive from one city to the next,” said DARPA Director Dr. Tony Tether in a DARPA press release. “After the success of this event, we believe the robotics community is ready to tackle vehicle operation inside city limits.”
Technology Development and Issue Management

Even as teams from across the United States prepare for the Urban Challenge 2007, numerous military organizations are asking for and awarding contracts to companies for their unmanned vehicle solutions. Although the military has implemented for tactical use a number of unmanned airplanes, relatively few land vehicles are currently in use, and none of them are completely autonomous.

The lack of autonomous ground vehicles is due in part to the complications of traveling across the Earth’s surface, which is different than traveling through the air. Whereas planes fly through a three dimensional (3D) space where all obstacles (other aircraft) are identified, land vehicles travel on a two dimensional (2D) plane where obstacles (think of them as disturbances in the 2D plane) may be unidentified and collision avoidance is essential. As we have noted above, however, safety in the air will become an increasing problem as the number of UAVs increases.

Humans rely primarily on vision to drive a vehicle, and other human senses allow them to travel a designated path and respond to obstacles by maneuvering around them. Unmanned vehicles must have also this ability to sense surroundings, travel a specified course and maneuver around obstacles.

The state of autonomy in unmanned vehicles is in its infancy. Currently, most unmanned vehicles are known as RPVs or Remotely Piloted Vehicles. All functions of an RPV are controlled by a human operator including propulsion and collision avoidance. Sometimes this is also referred to as tele-presence. RGV or Remotely Guided Vehicles are semi-autonomous. These vehicles have some ability to automatically react to observations of their surroundings. They are effectively still controlled by a remote operator, but do not require 100 percent attention. The Mars rovers are good examples of RGVs. The final class we’ll discuss are AGVs or Autonomously Guided Vehicles. These vehicles observe their surroundings and react to them based upon an “attitude”. They are provided a series of objectives and perform those objectives with little to no human input. That “attitude” gives them the level of aggressiveness that they will use to complete the objectives—wait at the stop light forever or run it.

The levels of autonomy can be thought of as a linear progression from today with RPVs as the norm to tomorrow where AGVs are common place.

Future Market and Growth Potential
Congress created an incredibly large, new market with its mandate to autonomize military vehicles. An estimated 360,000 military land vehicles are currently in use, and unmanned vehicle technology can have wide relevance in industrial and commercial applications.

Because this technology lies on the forward edge of innovation, no clear leader exists within the unmanned vehicle industry. But individuals, corporations and governments are taking steps to leverage the possibilities of this fledging opportunity.

For example, the Center for Autonomous Vehicle Applied Technology and Information (CAVÀTI) is a non-profit, Utah-based business coalition with a mission to advance autonomous vehicle technology by leveraging industry assets with university resources in the state of Utah. CAVÀTI was instrumental in the placement of autonomous systems within Utah’s new Economic Cluster Initiative.

Just as ASTRAEA has done in Wales, CAVÀTI has also recognized the potential for growth in the unmanned vehicle market and has established SWARM, a collaboration among organizations in Utah that are interested in the autonomous industry. SWARM includes four objectives that have been created for the purpose of securing a significant portion of the retrofit market and bringing business. These objectives include securing autonomous vehicle contracts, establishing a nationally recognized test facility, promoting legislative initiatives to support AGV business and technology, and fostering autonomous systems research. This research agenda is already underway. CAVÀTI is spearheading a statewide campaign to facilitate increased participation by Utah teams in the DARPA Urban Challenge 2007, and it has established Dugway Proving Ground as a test facility for small to medium autonomous vehicles (SMAV).

Other organizations are collaborating in similar ways. The Technology Collaborative, a Pittsburgh, Pennsylvania-based organization, has a goal to facilitate digital and robotic innovation by growing companies and jobs within Pittsburgh’s robotics industry.

All of these collaborations and innovations will lead down a similar path—using unmanned vehicles in military settings to take the driver out of the loop and save lives. Eventually, this technology will migrate to industrial and commercial uses. It is not unlikely that farmers will use this technology to plow fields, weed, fertilize, and harvest crops, or that consumers will have vehicles that drive them around town. Astronauts may very well have unmanned vehicles capable of fulfilling missions on the surface of the moon or other planets.
Conclusion

Both air and land unmanned vehicles are moving forward in a dynamic, some might say frantic, pace. The military applications are currently spurring this amazing growth. Eventually, civilian applications will become, perhaps, even larger in volume and complexity. In this field, our creative energies have just begun to be applied. The future for unmanned vehicles is exciting but unknown. The one certain thing with unmanned vehicle technology is that innovators within the industry can imagine uses that are beyond our current horizon.

Acknowledgements
Special thanks for assistance in the article to Dr. Sue Wolfe, Aerospace Strategy Manager, Welsh Assembly Government and to Peter Lippincott, Spring, O’Brien & Co.

Brent Strong, a professor at Brigham Young University, is a contributing editor of Composites Manufacturing.

Deryl Snyder is an assistant professor at Brigham Young University and was previously an Aerospace Engineer support contractor for the U.S. Air Force (Chief Engineer for the AFRL BATCAM technology demonstrator program).

Troy Takach is a senior managing partner of DesignJug. He has more than 20 years of senior management experience in the embedded systems industry.