Developing a see-and-avoid system that enables unmanned aircraft to operate safely alongside manned aircraft in civil airspace is the industry’s greatest quest

What will be the core technologies that allow unmanned air vehicle operators to comply with civil regulators’ requirements that UAVs demonstrate an “equivalent level of safety, comparable to see-and-avoid requirements for manned aircraft”?

This requirement poses one of the major technical hurdles still confronting the UAV community in its drive for open airspace access. Work on the enabling technology is at the forefront of UAV research and development activities around the globe. Finding the right answer represents one of the most potentially lucrative areas of high financial payoff for UAV sector companies in the near term if lightweight, easily integrated and technically reliable solutions can be developed.

Nick Sabatini, FAA associate administrator for aviation safety, acknowledges the criticality of the challenge: “While we may be very desirous and supportive of introducing unmanned aircraft, we need to do be sure that we have an equivalent level of safety. That is really a challenge for industry. They need to do the research and development to assure that detect, sense and avoid is a practical and realistic technology. There is no way we can accommodate UAVs in the US national airspace system, in the same airspace, unless we can be assured that that is going to be realised…The holy grail is going to be detect, sense and avoid.”

Going for goal

But it is unlikely that any one single technological solution will achieve that goal. John Timmerman, head of systems operations at the FAA, says even with an appropriate detection system, complexity increases at every step to encompass the full spectrum of UAV design and performance: “First you have got to detect that there is an intruder. You have got to track the aircraft that may be a factor to you. You have got to determine if there is a conflict potential. You have got to decide in an evasive manoeuvre is going to be required. And then you have got to initiate the manoeuvre.” The problem is not just a matter of an effective sensing system.

In turn, once a technical solution is avail­able, it must operate within the safety para­meters of current national and international air traffic management (ATM) regimes. That means “UAV operations should not increase the risk to other airspace users and should not deny the airspace to them”, says Wg Cdr Mike Strong, UK Royal Air Force, and member of Euro­control’s UAV operational air traffic task force.

The Eurocontrol task force was set up in late 2003 to advise on military UAV airspace integration specifications for potential adoption by European member states as a complement to the separate exploration of civil regulations by the former Joint Aviation Authorities. The task force finalised a set 25 draft specifications earlier this year, with this undergoing an independent safety audit during June and July. The final version of the specifications is now in preparation, with these to undergo review via the Eurocontrol consultative process from late this year.

The draft military specifications, Strong says, are based on the principle that ATM regulations and procedures for UAVs should “mirror as closely as possible those applicable to manned aircraft”. UAV-specific ATM procedures should only be implemented, he says, “where the absence of an onboard pilot, particularly in combination of loss with the control datalink, generates the need for special arrangements. Otherwise the provision of an air traffic service to a UAV should be transparent to the ATC controller and other airspace users.”

In that framework, collision avoidance is first and foremost a function of air traffic control, says Strong. “However, when ATC is not available to separate a UAV from other airspace users, the pilot in command will assume this responsibility using available surveillance information and technical assistance in the form of a sense-and-avoid system.”

The performance parameters of that sense-and-avoid capability, Strong says, need to be based on internationally agreed standards, including minimum separation requirements. Given that there is no agreed consensus on this requirement, the draft Eurocontrol specifications assume that “more than one minimum separation distance is required and that a layered application, rather like an onion skin, would be more appropriate.

“The task force considered that where collision avoidance is provided by ATC within controlled airspace, the separation minima between UAVs operating instrument flight rules (IFR) and other IFR traffic should be at least the same as for manned aircraft flying in the same class of airspace.”

Collision avoidance

In circumstances where responsibility for maintaining safe separation rests with the UAV system and the unmanned aircraft remains under pilot control, the Eurocontrol taskforce is proposing a minimum separation distance of 500ft (150m). The requirement for fully autonomous collision avoidance being performed by a sense-and-avoid system could be less stringent, Strong says, with the task force concluding that it was “a last ditch tool akin to TCAS 2”. As such, the draft proposes that UAVs operating under the control of a sense-and-avoid system should be required “to achieve miss distances similar to those designed into ACAS [airborne collision avoidance system]. The system should be compatible with, and not compromise the operation of, ACAS.”

But Strong also cautions that this draft requirement does not mean that existing ACAS technology can be considered a ready solution to the UAV problem. He points out that International Civil Aviation Organisation “is opposed to any such fitment at present. It argues that operation of ACAS is predicated on having a pilot in the cockpit and also fitting ACAS to UAVs may be counter-productive in terms of reducing collision risk.”

According to Timmerman, UAV-specific sense-and-avoid systems must be based on mature and robust technologies if they are to meet the equivalent level of safety policy provisions. A baseline solution that would safely allow for UAV operations in visual flight rules conditions, he cautions, “is more than five to eight years away according to our experts from [FAA] flight standards and aircraft certification”.

Not all UAV users are prepared to accept waiting that long, however, with a variety of organisations exploring options for interim solutions for specific UAV classes in lieu of a permanent solution.

Despite the reservations that others may have about the adaptation of ACAS, Access 5 regards its looming demonstrations of ADS-B surveillance datalink and TCAS 2 collision avoidance using the Scaled Composites Proteus as an important part of “confidence building”. Access 5 strategic communications manager Dale Tietz says those flights will explore both “command and control, and sense-and-avoid”.

The US Coast Guard aims to bring its Bell Helicopter Eagle Eye tilt rotor UAVs into initial operational service from 2007 with sense-and-avoid capabilities provided by the Telephonics RDR 1700CG X-band radar. That system is expected to be able to detect airborne objects with a radar cross-section of 1m2 (11ft2) at ranges of 9km (5km) and objects with a cross-section of 5m2 at 18.5km. Telephonics says the first production units will be available for initial integration into Eagle Eyes in 2006 and will undergo specific sense and recognise and sense-and-avoid testing in 2007.

Demonstration relaunch

In June this year the US Air Force’s UAV Battlelab relaunched its air traffic integration programme with the issuing of a new broad agency announcement seeking proposals for the “demonstration of a mature technology” that can “perform the sense-and-avoid function on a Predator-class aircraft. The system must have at least 110° coverage right and left of the centreline and ±15° of flight path.”

The relaunch follows the acknowledged failure of a prototype optical system during Battlelab trials late last year. That system, developed by Defence Research Associates and the US Air Force Research Laboratories sensors directorate, comprised three optical sensors linked to a processor to calculate the closure rate of detected objects. Intended as a near-term sense-and-avoid solution for USAF MQ-1 Predator UAVs, the system was, says Colonel Larry Felder, commander of the USAF UAV Battlelab, “a failure. It did not work. We had multiple false targets.”



Source: Flight International