Civilian drones

WHEN drones are used even by environmental activists to track down Japanese whaling vessels, it is a sure sign that UAVs (Unmanned Aerial Vehicles) are no longer the sole prerogative of the military. Police forces around the world are certainly keen to lay their hands on small pilotless aircraft to help them nab fleeing criminals and monitor crime scenes from above. With price tags of a little more (and, in some case, a good deal less) than the $40,000 of a patrol car, a new generation of micro-UAVs is being recruited to replace police helicopters costing $1.7m and up.

And the police are not the only ones eager to take advantage of the technology developed for attacking terrorist hideouts in Afghanistan and Pakistan. Any civilian activity that would be improved by having an aerial view—monitoring traffic, checking electricity cables and pipelines, surveying forestry and crops, taking aerial photographs, patrolling wooded areas for fire—could benefit from the use of UAVs.

The widespread use of such drones, though, raises questions. Some are of safety: every extra craft in the air adds to the risk of a crash or collision. Others are of privacy: are people's activities to be monitored continuously when they are outdoors, even when they are on their own, private property? In America, in particular, these questions are at least being debated.

The Federal Aviation Administration (FAA), for instance, has issued some 285 temporary permits for testing UAVs in airspace where commercial air traffic and private aircraft operate. But because none of the UAVs intended for civilian use can yet comply with the FAA’s “sense and avoid” rules, a ground observer or chase aircraft has to keep the drone in sight at all times to act as its eyes.

The FAA is working on regulations that will allow unmanned aircraft to operate routinely in public airspace without creating a hazard for other air traffic and people on the ground. The agency was due to issue its draft this January. Unfortunately, political squabbles in Congress delayed the FAA’s budget, and the agency had to shut its doors for a while. The forced furlough has meant the new regulations will not be available until spring.

One technical issue the FAA has been addressing is how a UAV should respond if it loses its communications link with the operator on the ground. Should it automatically return to some pre-assigned GPS location, or head for the nearest open space? Should it have a parachute arrangement—like an increasing number of private planes—to lower it gently to the ground in an emergency, or should it put itself immediately into a stall?

Plenty of practical solutions exist for such problems. The only issue is cost. A bigger stumbling block is how UAVs should detect, sense and avoid other aircraft operating in the same airspace. What bothers the FAA is that drones piloted remotely by operators on the ground cannot see other aircraft in the sky in the way that human pilots can. Before giving the go-ahead, the agency wants UAVs to be able to operate as safely as manned aircraft. That means developing a lot of expensive gear to avoid mid-air collisions and near misses.

The armed services, which are also keen to use their drones on training missions in unrestricted airspace, think the FAA’s concerns are overblown. They want to be able simply to file a flight plan with local air-traffic controllers and then send their drones aloft. They argue that, thanks to visual and infrared sensors capable of resolving objects five miles ahead in great detail, the current generation of military UAVs can see air traffic better than most human pilots can, and take even swifter evasive action if necessary.

That may be true for sophisticated UAVs like the Predator that have two separate sensor "balls"—one surveying the ground, the other scanning the surrounding airspace—each with its own human operator on the ground. But not everything is that well equipped. Besides, seeing is one thing; being seen is an entirely different matter. Like parafoils and hang-gliders, most micro-UAVs are made of composite materials that leave only the faintest of radar footprints. Even some of the bigger military UAVs are not all that easy to spot in the sky.

Last August, a C-130 Hercules freighter plowed into a military drone while coming into land at a base in eastern Afghanistan. By all accounts, the UAV was where it was supposed to be—circling 4,500ft above the end of the runway, awaiting permission to land. The C-130’s flight crew apparently neither saw it, nor were they on the correct flight path. That was not the first such incident in Afghanistan.

Such mid-air collisions could have been averted if what the forces call a “deconfliction” system had been in place. Ironically, the army at the time was testing a sense-and-avoid prototype at a training centre in southern California. The prototype, built by General Atomics, of San Diego, uses radars at three airports in the region to triangulate and track all aircraft entering a given volume of airspace. When approved by the FAA, the army’s sense-and-avoid system will allow its UAVs to fly in and out of public airspace.

Research groups around the world are seeking cheaper solutions to the deconfliction problem—especially for micro-UAVs. One of the more promising approaches uses an array of lightweight acoustical probes coupled to a signal-processor that filters out wind noise to listen for and locate other aircraft in the sky, and to transmit avoidance instructions immediately to the drone’s operator on the ground. One such sense-and-avoid system, developed  by SARA, a contract-research firm based in Cypress, California, weighs little more than eight ounces.

Clearly, the police and other civilian groups are unlikely to be able to afford having a couple of dedicated operators on the ground for each UAV in the air. Nor can they be expected to have multiple ground-based radars tracking their UAV’s every move. Tiny sense-and avoid systems like SARA’s passive acoustical array should help minimise conflicts in the air. But if small pilotless planes are to fulfill a useful role in fighting crime and saving lives, then the regulations governing their use ought to reflect the environment in which they are likely to operate.

Because of the roles they will play, most micro-UAVs used by civilian agencies will operate well below 400ft and probably (like the rules governing model aircraft) no closer than three miles from an airport. That is not exactly Class A airspace used by commercial air traffic. Mid-air collisions are therefore even less likely than they are in the open skies.

Size also matters. The majority of pilotless planes that civilian agencies have their eyes on are little bigger than model aircraft and weigh much the same. One that is fancied by America’s 18,000 state and local police forces is a pilotless helicopter called the Qube. This four-rotor craft, made by AeroVironment of Monrovia, California, weighs in at 5.5lb (2.5kg) and fits easily in the boot of a car. It can be assembled in minutes to provide an immediate eye in the sky capable of staying aloft for 40 minutes at a time. With its battery powered electric motor, the three-foot long Qube makes barely a whisper.

AeroVironment is America’s leading supplier of micro-UAVs to the armed forces, having delivered more than 20,000 new and replacement craft to forces at home and abroad. Apart from its Qube, the company’s family of small UAVs include the Wasp, Raven and Puma, ranging in weight from 1lb to 13lb. All are carried on the battlefield in backpacks, assembled in minutes and launched by hand. All feature high-resolution colour and infra-red cameras for streaming live video back to a soldier on the ground using a hand-held controller resembling a video-game console.

With their versatility, low price and modest operating cost, micro-UAVs like those made by AeroVironment are expected to be the most widely adopted by civil and commercial operators. With most weighing less than 10lb, they have much the same kinetic energy as a large bird. In other words, the threat they pose to other planes in the sky and property on the ground is akin to a bird strike.

That can be serious, of course. Flying into a flock of birds can bring a jetplane down—as happened most famously to Capt Chesley Sullenberger’s Airbus 320 after taking off from LaGuardia airport in New York, in January 2009. But airlines live with the threat of bird strikes every day. So do people on the ground. Your correspondent is forever burying dead birds that have crashed into his shaded ceiling-to-floor windows and replacing damaged window screens.

But the biggest issue surrounding the domestic use of drones could be not technical but legal. In America, at least, neither the Constitution nor common law prohibits the police, the media or anyone else from operating surveillance drones in the neighborhood. As the law stands, citizens do not have a reasonable expectation of privacy in a public place. And that includes parts of their own back yards that are visible from a public vantage point, including the sky. The Supreme Court has been very clear on the matter.

In theory, then, there should be little cause for debate over the introduction of micro-UAVs. There have been surprisingly few complaints, for instance, about technologies such as biometrics, facial recognition, RFID tags or GPS location—all of which pose as much of a threat to personal privacy. So, why should civilian drones be any different?

In a thoughtful essay published in the Stanford Law Review last month, Ryan Calo of Stanford Law School argues that things could be indeed different this time. “Virtually any robot can engender a certain amount of discomfort, let alone one associated in the mind of the average American with spy operations or targeted killing,” says Mr Calo.

Unlike, say, the National Security Agency’s surveillance network or commercial data brokerages that function secretively in the background, surveillance by civilian authorities is likely to be highly visible. “People would feel observed, how or whether the information was actually used,” Mr Calo notes. The resulting backlash could force the courts, legal scholars and civil-rights activist to re-examine not merely the use of drones, but the actual doctrines that permit their use today.