Thursday, June 1, 2006
Europe's Vision of Future Avionics
Avionics will change in all its dimensions between now and 2020. We take a look into the future, based on a recent Eurocontrol study.
At A Glance:
As part of a wider issue focusing on the future Overall ATM/communications, navigation and surveillance (CNS) Target Architecture (OATA), Eurocontrol has released an avionics study targeting three time frames: now-to-2007, 2007-2011 and 2011-2020
In today's air traffic management (ATM) model, aircraft follow fixed routes defined by ground navaids, surveillance depends on ground radar, communication is via voice, and ground controllers call the shots. But this model is no longer adequate to meet future needs. It uses controller resources and airspace inefficiently and will eventually, in congested areas, run out of capacity. The international aviation community has proposed answers. It wants to shift some of the responsibility for deciding and controlling aircraft trajectories back to the cockpit, where it resided in the early days of flying. This would simplify the ground control task and free controllers to handle more traffic. And it would release aircraft from the confines of present "aerial roadways" to become "off-roaders," flying more direct routes to their destinations and making better use of available airspace.
Around airports, the aviation community wants aircraft to follow planned trajectories precisely, so that they can safely be allowed to fly closer to each other and to ground hazards. Controlling trajectories in time as well as in the three spatial dimensions will further improve the situation. And extending the new control regime to embrace aircraft movements from gate to gate--including on the ground--will be another lynch pin. All these measures should help make management of aviation traffic more efficient, save time and fuel, and reduce environmental impacts. But how will it affect avionics and the way the flight deck will look in, say, 2020?
Europe's senior ATM body, Eurocontrol, has painted a picture of anticipated developments in an avionics study that is part of a wider deliberation addressing the future Overall ATM/communications, navigation and surveillance (CNS) Target Architecture (OATA). In fact the OATA study paints pictures for three time frames: now-to-2007, 2007-2011 and 2011-2020.
Europe has a particular problem: its already dense air traffic is expected to double by 2020, according to Alex Hendriks, head of Airspace/Flow Management & Navigation at Eurocontrol. The current onboard and ground-based infrastructure, he adds, will not long be able to cope with such growth. Fortunately, compact computing, data links, augmented space-based navigation and other evolving technologies offer solutions. In common with the International Civil Aviation Organization (ICAO), FAA, Boeing, the U.S. Air Transport Association (ATA) and other interested bodies, Eurocontrol believes that such technology, along with a more balanced sharing of ATM responsibility between ground and air, can solve the capacity problem. The challenge, though, is to achieve an orderly transition in harmony with other regions, so that solutions can be global. The difficulty is exacerbated by the need to allow for legacy systems, since 60 percent of aircraft now flying are still expected to be in service in 2020, according to the study.
The outlines of the eventual solution seem clear. Satellite-based navigation would largely replace ground-based navaids. Communications would become digital and data link-based. Having aircraft report their own positions and associated parameters would enhance surveillance and reduce the need for expensive ground radar. "Free routing," or "free flight," based on virtual waypoints, would permit fuller use of available airspace. Information would be more predictive and would be shared by all involved parties. All this would enable air crews to have a greater say in determining journey trajectories, which would be negotiated rather than prescribed. Some responsibility for ensuring aircraft separations, along with sequencing and merging in the approach, may be devolved to the flight deck. Aircraft could even operate autonomously in non-core areas.
Under Europe's Single-Sky vision, airspace will be managed in large functional blocks, reducing the present need for repeated handoffs between today's smaller sectors and making better use of controller resources. Military aviators will rely less on airspace permanently assigned to them but will benefit from sharing efficiently managed airspace with civil users.
Automation is expected to affect every level, from controller-pilot data link communication (CPDLC) to advanced onboard flight management systems (FMS) controlling entire 4D flight path trajectories. The present tactical approach to ATM, in which controllers intervene as traffic conflicts and other hazards become apparent during flight, will give way to a more strategic approach in which hazards are foreseen and allowed for at the flight planning stage. Wim Post, Eurocontrol's activity manager, Research Coordination and Validation, summarizes the required management style as "less hands-on and reactive--more proactive."
Obstacles to this vision are more institutional and economic than technological. In fact, aviation may have a surfeit of technology options. This has prompted a transition, now in progress, from specifying avionics equipment standards to the technology-agnostic approach of specifying required performance instead. Any technology able to provide the stated performance--required navigation performance (RNP) is an early example--can be allowed, including legacy equipment. This performance-led approach reduces the tendency, noted in the past, for solutions to be technology-led. As Hendricks commented at a joint Eurocontrol/FAA-sponsored avionics workshop held in Toulouse last year, "Technology for technology's sake is of no interest; operational requirements must be the driver."
The latest Airbus and Boeing aircraft are equipped at delivery with avionics able to meet evolving performance standards, including RNP. Operators complain, however, that they cannot use the full capabilities of their avionics because ground equipage has not kept up. Persuading multiple air navigation service providers (ANSPs) to upgrade air traffic control (ATC) centers in step with each other and with user requirements is a major part of the challenge facing Eurocontrol. Given the diversity of national interests and budgets across Europe, it will be difficult.
Eurocontrol expects that present federated avionics architectures eventually will give way to digital integrated modular avionics (IMA). It makes little sense to have up to 50 individual systems and 100 processor boxes on a large contemporary jet, when a single powerful computing resource could be shared by many functions. IMA reduces the number of subsystem line replaceable units (LRUs), each with its own chassis, power supply and processor, thereby cutting aircraft weight, complexity and certification costs. A data network enables subsystem modules to communicate with each other, and new modules can be added on a plug-and-play basis when alterations or upgrades occur. Open standards and common software make additions more manufacturer-independent. IMA standards are evolving, and the avionics community is homing in on a standard common avionics computer resource (or computer) as so far defined in RTCA DO-255, plus ARINC's specification 653 for the software.
Transitioning to such logical architectures is a long-term aim, even though they already are appearing on the latest Airbus and Boeing transports and some business jets. IMA and federated systems will operate alongside each other, with the balance gradually shifting to integrated avionics.
Developments already in progress are key to the immediate future. A start has been made on digitizing and automating routine controller-pilot communications in en-route continental airspace and terminal areas, and progress continues under Eurocontrol's ongoing Link 2000+ and CASCADE programs (April 2006, page 28). This has already enhanced safety and reduced delays by decreasing the time spent in fallible voice procedures. Analog VHF voice will nevertheless remain dominant for some time to come, and VHF channel congestion may be further addressed by extending mandatory 8.33-KHz channel spacing to the majority of European airspace, vertically and horizontally. Most aircraft now carry 8.33-KHz-capable VHF transceivers.
The present airborne communications addressing and reporting system (ACARS) data link will continue to be used in this time frame, according to the Eurocontrol study, but the higher-speed VHF digital link, Mode 2 (VDL-2)/aeronautical telecommunications network (ATN) technology now being introduced will greatly increase data link capacity. The corresponding ground infrastructure, already present at Europe's major en-route control center at Maastricht in the Netherlands, will be extended to key area control centers. Aircraft in oceanic airspace can use satcom, HF radio (via HF data link communications, or HFDL) or ACARS with future air navigation system (FANS) avionics.
Europe continues to improve ground surveillance with new radars, in particular secondary surveillance radars (SSRs). But the region also is starting to implement the complementary automatic dependant surveillance-broadcast (ADS-B) system, in which aircraft downlink their own satellite-derived positions via the SSR-associated Mode S data link. ADS-B, which requires participating aircraft to be fitted with transponders, can be used to complement radar or replace it in areas without radar coverage. Its accuracy remains constant regardless of distance.
ADS-B also enables aircraft to downlink a range of other parameters useful to controllers, such as airspeed, ground speed, heading, selected flight level and vertical rate. This information is extracted from the aircraft's flight management system. The extended squitter (ES) version of the Mode S data link has the capacity to downlink other information too, such as weather observations and, eventually, intent data. Knowledge of aircraft intent (via selected waypoints) can be used to improve the accuracy of trajectory prediction. And trajectory prediction can be further enhanced by allowing for likely en-route weather, forecast using a meteorological model built up from aggregated aircraft weather observations.
Not all airborne parameters need go to controllers--it would be easy to overload them. Some can go to other departments, such as airline operations, so that monitoring tools, safety nets, alert systems and other ground functions can be developed. Data also can be linked to other aircraft so that they can assemble their own "pictures" of proximate traffic. Importantly, too, ADS-B can facilitate surveillance of traffic on the airport surface, contributing to the more advanced surface movement guidance and control systems (A-SMGCS) expected to be implemented at leading airports in the near term. (In this context, the term "traffic" refers to all aircraft and ground vehicles fitted with the necessary transponder/GPS combination to be visible to controllers.) There is also some support for multilateration, an alternative tracking system for aircraft and other assets at short and medium ranges. Providing air crews, controllers and vehicle drivers with moving-map traffic situation displays will promote safe operations on the airport surface, especially in reduced visibility conditions. It should also reduce runway incursions.
On the navigation front, Europe's steady adoption of basic and precision area navigation (BRNAV and PRNAV) will continue during this period, though it will coexist with conventional ground-based navaids. BRNAV has been mandatory for en-route airspace in Europe since 1998, while PRNAV is already in use in some terminal airspace. A key enabler is global navigation satellite systems (GNSS), which at present consist of GPS and GLONASS. RNP, the performance-based successor to the BRNAV/PRNAV concept, will not progress significantly in Europe until after 2007 because of procedures and equipage issues.
Inertial navigation and systems based on DME are being further developed as backup positioning sources, given the now general acceptance that GNSS cannot be adequate as a sole means of navigation. Loran is another possible backup, though the OATA study leans towards DME/DME at present. Airborne collision avoidance systems (ACAS) plus terrain awareness warning systems (TAWS) will remain mandated as last-ditch protection against collisions with other aircraft or the ground. Microwave landing systems (MLS) might be installed at some airports where conventional ILS systems are precluded.
Notwithstanding these developments, the OATA study envisages that air traffic control will stay mainly tactical and ground-based during this period. LRU-based avionics will still predominate.
This period is a bridge between the current tactical ATM system and the more strategic system of the future, based on trajectory exchange and optimization in a cooperative environment. Early examples of trajectory negotiation may be seen during the period.
New data services will be implemented, using VDL-2 or other ATN-compliant high-speed, air-ground data links, according to the report. Ground-based air traffic service units (ATSUs), or facilities, will be able to access information contained in the flight management systems of target aircraft. They will be able, for example, to retrieve route data and check its consistency with filed flight plans, or gain knowledge of aircraft intent to assist in conflict detection and resolution. Any trajectory changes thought necessary can then be negotiated. Pilots will be able to downlink their preferences to ATSUs so that controllers can allow for them when proposing actions. Some functions will be automatic. For instance, an aircraft may downlink trajectory data to an ATSU when triggered by a specified set of criteria.
Combining downlinked aircraft parameters with enhanced surveillance will enable new ATM tools to be developed. Arrival/departure managers and ground-based safety nets are examples. More aircraft parameters may be needed. Weight, engine variant and airline policy on thrust settings, for example, are factors that can influence trajectories so, if ATSUs were to be made aware of them, they could provide more accurate trajectory predictions. This would help in collaborative determination of trajectories.
More operational flight information--such as weather, NOTAMs and conditions at destination airports--will be uplinked to aircraft. Pilots will be able to spend less time transcribing voice reports and more managing the flight. The latter may include interacting with ATSUs, via data link, in traffic conflict avoidance tasks. Pilots may be empowered to do this by purpose-engineered cockpit displays, electronic flight bags or other displays (e.g., TCAS) that show proximate traffic with aircraft identities, the report states. ATM benefits will depend on the proportion of aircraft in the operational area that are ADS-B equipped. But Eurocontrol expects that a significant percentage of airplanes within this period will be fitted at least to transmit ADS-B information ("ADS-B out"), even though it will take longer for them to be equipped to receive the ADS-B information from other aircraft ("ADS-B in") that will be necessary for full airborne surveillance.
These technologies will pave the way for possible collaboration between the air and ground sides on separation assurance, as a prelude to more extensive transfer of responsibility to the flight deck. However, the desirability and extent of such radical role changes are still debated. Some experts question whether the result might not be too "democratic," compromising efficiency and safety, while others say shared responsibility is essential to progress. Pending final commitment to this "devolution," pilots might at least be empowered to maintain their own separations as instructed by ground controllers, an intermediate step that still offers significant capacity benefits.
More complex spacing tasks, such as en-route establishment of in-trail and level spacings, level crossing and passing, may be tried experimentally during this period. Again, ground controllers would advise pilots of the spacings required. Once advised, spacings can be maintained precisely--along with high-level automation involving coupling of GNSS, FMS and autopilot system--it would be possible to reduce separation standards, increasing airspace capacity.
Participating aircraft will need avionics of the highest integrity, including an ADS-B receiver, a cockpit display of traffic information (CDTI), an airborne separation assistance system (ASAS) processor and a system delivering RNP, according to the report. Much work is required to define the necessary standards. Eurocontrol and FAA are, for instance, supporting work to determine standards and requirements for the cockpit display device.
On the airport surface, A-SMGCS will make it feasible for controllers to route aircraft and ground vehicles optimally, using automated support tools. Taxi instructions will be delivered via data link to cockpit displays, or to onboard computers. Giving pilots and drivers a picture of the surface situation superimposed over a moving map will enhance the safety of surface operations and reduce delays in impaired visibility. The ground system should, it is envisaged, be able to manage ground lighting (stop bars, alarms, etc.) according to cleared taxi routes. Adherence to planned routes will be monitored and alerts given of potential runway incursions. Taxi times and runway occupancy times should be reduced.
Transition to RNP
Meanwhile, the transition to RNP airspace will continue. More RNP procedures will be introduced, enabling pilots to opt for preferred routings and to reroute dynamically during flight. Navigational precision and system integrity will gain from the progressive introduction of Europe's Galileo GNSS, complementing GPS. The enhanced integrity and improved vertical navigation capabilities of such combined GNSS, with ground or space-based augmentation, will permit the introduction of precision approaches to at least Cat II standards, according to the report. This would be a starting point for GNSS landing systems (GLS), which are expected to save on ILS infrastructure costs in the long run.
European airspace, simplified to just three categories, would become more adaptable, responding quickly to changes in circumstances or user intention, the study predicts. (The three airspace categories are "N"--airspace in which all flights and their intentions are known; "K"--airspace in which all flights, but not their intentions, are known; and "U"--uncontrolled airspace.)
RNP will permit the shrinking of obstacle protection areas and the avoidance of environmentally sensitive areas. Air traffic controllers would benefit from new RNAV applications such as "parallel offset," "direct to" methods and separation tools. Precision flight path following will make it easier to arrange "clean" climbs and descents and to comply with noise abatement strictures without seriously degrading airport capacity.
RNP compliance will require close coupling between the automatic flight control system (autopilot) and the flight management system. Automatic control is needed because human pilots will be unable to follow the flight trajectories stored in the FMS with sufficient reliability and precision. Reduced pilot workload at critical flight stages, notably in terminal areas, will be an added safety benefit. Eurocontrol accepts that more work is needed on integration and interoperability, the consistency of navigational data and the surveillance picture between air and ground, and the need for ground controllers to be advised of the RNP status of particular aircraft.
The OATA study predicts that by 2011 momentum will be building behind the new avionics, as the number of equipped aircraft increases and airspace users become convinced of the benefits. Even so, traditional architectures still will be present in the majority of aircraft.
By 2020, however, the situation will have evolved further. New CNS/ATM avionics should be dominant by that time, and Eurocontrol expects to have moved closer to the vision outlined at the beginning of this article. Strategic ATM based on 4D trajectory exchange and optimization is likely to begin during this period. The advent of 4D trajectories to adjust flight plans before airplane departure will help flight planners avoid air traffic system overloads. Rationalized airspace would transcend national boundaries, require fewer ATC centers and support adoption of more predictive airborne and ground-based systems. Free routing would apply to most ECAC airspace, according to the report. Sharing of a common information pool by all stakeholders would facilitate collaborative decision making. Machines would "talk" to each other, leaving pilots and controllers to communicate only in the case of deviations from plan. Tactical ATC intervention would be the exception rather than the rule.
Digital data link coms would make flight planning, modification and coordination a continuous process. Changes in flight trajectory would be communicated directly to onboard FMS via secure data links. Users will be able to negotiate preferred trajectories, reducing delays for 4D-equipped aircraft. In terminal areas 4D RNAV would permit consistent and closer spacing, ensure efficient timing and assist accurate approach sequencing. Within this period, some of the airborne separation assurance task is expected to be transferred to the flight deck and autonomous operations may be undertaken in non-core European airspace.
Improved surface operations, based on A-SMGCS and a common surveillance picture, as already described, are likely to become general, the report predicts.
A substantial percentage of aircraft will have been equipped with flexible IMA-based avionics having high software content. It therefore will be possible to add functionality by upgrading software rather than by having to add LRUs and wiring. It will become feasible to upgrade legacy aircraft with some IMA elements so that operators can reap some of the benefits. The avionics retrofit business will gain from this. Stringent certification requirements will, however, inflate the cost of new and upgraded equipment, prompting the OATA study authors to caution that implementing the new regime will be a progressive rather than a "big bang" affair.
Integrated Modular Avionics: B787
Perhaps the most ambitious implementation of integrated modular avionics (IMA) so far is the Boeing 787 Common Core System (CCS), the nerve center of the aircraft. Designed by Smiths Aerospace, CCS supports nearly 100 applications. The system consists of processing resources; the common data network, which provides high-speed communications between processors and function-specific avionics and utilities equipment; and remote data concentrators, which enable any system to connect with analog, discrete and other legacy interfaces over the core network.
“The Common Core System is very true to the vision of IMA,” says Mike Sinnett, Boeing’s chief engineer for systems with the 787 program. CCS, he explains, embodies the IMA concepts of modularity, openness and scalability. It features “common computing elements that operate in an open architecture–across an industry-standard interface between hardware and software–on a common data network serving the bulk of the computing needs of the entire airplane.” Smiths currently is shipping prototype CCS devices to suppliers of software hosted on the system. Sinnett wants CCS to be “service ready” before flight tests begin next summer.
Boeing took this approach in order to reduce weight and volume, development cost and the future cost of change. Sinnett expects the B787 will be in service for 60 years after the last aircraft is built, during which time there will be many changes that aren’t even imagined yet. The airframer already is planning a CCS technology roadmap for processor upgrades.
The B787 goes far beyond Boeing’s next-most-integrated airplane, the B777. The 777’s aircraft information management system (AIMS) integrates approximately 16 previously standalone functions, says Sinnett. Although the Honeywell-designed AIMS was a breakthrough for its time, 70 to 90 different types of line replaceable units, or LRUs, remain, a number of which require multiple copies. Each LRU type requires not only its own software, but also its own backplane, processor, memory and power supply. Each box also requires wiring to connect to other systems. The 787’s CCS, by contrast, supports about 90 functions and requires only about 30 different types of LRUs. The 787’s architecture will save hundreds of pounds in wiring alone, compared with a federated implementation, according to Boeing.
The 787’s avionics architecture is also more open, and Boeing is counting on this openness to minimize the cost of evolution. CCS is based on ARINC 653 for partitioning operating systems–a foundation of IMA–ARINC 664 for data networking and DO-178B, Level A, for safety-critical software, plus other industry interfaces such as CANbus and ARINC 429. Boeing also has captured all of the interface data in one interface control document for 787 suppliers. “The big secret of the open architecture is that it’s no secret,” Sinnett says.
Changing a processor is more difficult in a more federated system because the hardware and software are so tightly intertwined. “If you change the processor, you have to recertify the system and all aspects of the LRU,” as well as determine the impacts on associated systems, Sinnett says. If the change involves the autoland system, for example, “you’d likely have to do a new autoland certification, which means a lot of flight time and time with the regulatory agencies, and that can get very expensive.” With an open architecture, however, “you abstract the software from the hardware on which it runs,” he says. This is designed to make it possible to change the hardware while limiting the impact on the software.
IMA is moving toward a different certification paradigm, predicts Mike Madden, Smiths’ 787 program director. “We’re headed toward certifying applications independently of platforms.” This includes migrating certified software from one aircraft to another–without recertifying all the software’s functions in the new platform–as well as adding new applications to the original platform. Just how such “modular certification” will work out in practice is the subject of standards activity and regulatory debate, but Smiths and Boeing are confident of the outcome.
The CCS “computer,” as currently configured, comprises two cabinets, each loaded with the same 16 modules–in five different types–but each running different combinations of software applications. Boeing currently plans to go into flight tests with a full-up cabinet, including eight processing modules, two (dual-redundant) power control modules, two network switches, two fiber optic translator modules and two graphics generator modules. Smiths uses Wind River Systems’ “Platform for Safety Critical ARINC 653” commercial real-time operating system (RTOS) as the core software, allowing multiple applications to share the same hardware. The RTOS theoretically can support up to 256 partitions, or functions, although it is unlikely to be called upon to do so.
Of the CCS’ 90 applications, about 46 unique ones are processed mainly within the cabinets, Madden says. The common data network includes both a 100-Mbit/s fiber optic “trunk” line and 10-Mbit/s copper wires, both built to the ARINC 664 aviation Ethernet standard. Among the major CCS functions are flight management–implemented completely in software–navigation, displays, health management, crew alerting and maintenance. –Charlotte Adams