Enhanced vision systems are quickly entering service and synthetic vision is under test: will they form the perfect marriage?
Charles Lindbergh would have loved enhanced and synthetic vision systems (EVS/SVS). The sensor-generated, real-world EVS view would have guided the famed aviator to land in darkness or poor weather, or both. The artificial world view created by SVS would have compensated for the non-existent forward visibility from the cockpit of his Ryan NYP monoplane.
Yet Lindbergh foresaw the need and, having neither technology, proclaimed that "aviation will never amount to much unless we learn to free ourselves from the mists". Little wonder that his warning, along with his forlorn wish for "a pair of spectacles to see through the fog", are favourite quotes of staff at Rockwell Collins Advanced Technology Center. The team is one of several looking at combining EVS and SVS into one system that provides the best of both worlds: improved safety as well as real operational benefits - a vital driver towards wider adoption and development.
Today's efforts build on the pioneering work undertaken by Gulfstream, which, along with Kollsman and Honeywell, became the first to receive US Federal Aviation Administration certification for an EVS in 2001 on the GV. Since then, Gulfstream says, more than 100 EVS-equipped aircraft have entered service, including retrofits and new-delivery G550s, where EVS is standard equipment. The system, which projects infrared imagery from by a nose-mounted Kollsman 1-5µm cooled single-channel forward-looking infrared (FLIR) on to a Honeywell 2020 head-up-display (HUD) and a co-pilot auxiliary head-down display (HDD), will also be standard on the forthcoming G450.
Sales of this and other EVS systems under development by CMC Electronics, Max-Viz and Thales are being stimulated by an FAA rule passed in January allowing the first landing credit for EVS-equipped aircraft. The enhanced flight vision system (EFVS) operational rule allows pilots to continue a straight-in approach past decision height or minimum descent altitude down to 100ft (30m), provided the airport environment can be seen using the EVS. This represents a reduction in minima ranging from 100ft for a Category I landing to several hundred feet for non-precision approaches, and provides a Cat II-equivalent minima at 900ft-plus Cat I approaches. The rule allows use of any sensor providing a real-time image of the outside world on a HUD.
While all the initial systems, such as the Gulfstream EVS, use IR cameras, future developments are planned to include millimetre-wave (MMW) radar, together with enhanced IR and low-light television systems. Yet to achieve maximum "blind landing" credit potential from the FAA and European authorities, most developers recognise that some sort of integration with the computer-generated database of an SVS will be necessary, possibly augmented by other situational awareness systems.
Steps towards this vision were taken in this year during tests of a modified EVS-equipped GV around Reno, Nevada by a NASA, US Air Force Research Laboratory (AFRL) and industry team for the third and final phase of a five-year synthetic vision programme. Earlier phases, using NASA's Boeing 757 testbed, looked at situational-awareness benefits of SVS technologies that had originally been studied as part of the requirements for an "external vision system" for the High Speed Research programme. The supersonic effort died, but the technology work on sensors, terrain databases and sensor fusion and integration paved the way for the current studies that now fall beneath the umbrella of NASA's Aviation Safety and Security Programme.
More work needed
The Reno-based tests were conducted from June to August in day and night visual meteorological conditions and evaluated the use of an integrated EVS/SVS within the tight performance confines of approach procedures similar to those required for RNP (required navigation performance). "We looked at symbology declutter, and 'highway in the sky' or tunnel-type displays," says NASA Johnson Space Center-based SVS test pilot Rob Rivers. "It included how you fly through the tunnel and workload issues. There's a lot more work to do on this."
The evaluation included a HUD (a video-capable Rockwell Collins Flight Dynamics HGS-4000 in place of the standard Honeywell 2020), with the computer-generated terrain overlaid, and a hybrid display on the primary flight display. Although originally tested in full photo-realistic mode, Rivers says: "The problem is if a picture is taken in winter, when there's no vegetation, the terrain may look different in summer. We have found a photo-based generic hybrid image mixed with the database is best."
Related simulator tests of the hybrid display against a standard blue sky/brown ground navigation display and terrain awareness and warning system (TAWS) showed a "big preference for the hybrid display" among a group of 16 experienced airline pilots. Tasked with a tricky approach to a mountainous airport in a large aircraft with one engine failed, and the TAWS audio disabled, the test was to see "if the information available was enough" to prevent a controlled flight into terrain (CFIT). Prevention of CFIT is a primary mission of NASA's SVS research. All 12 of the pilots using the advanced EVS/SVS display "recognised and avoided" the hazard, while all four using the conventional display crashed. "We didn't think we'd have 100% correlation, but we did," says Rivers.
"One of the things we found out was a database error of 200ft that the pilot would not have detected until at least 500ft above that point. As a result, we think some sort of auto-detection warning system will be essential." The system would probably be a version of the DIME (database integrity measuring equipment) first evaluated during the second phase of the NASA tests in Eagle Vail, Colorado in 2001.
Integrity, and how to keep the pilots constantly aware of the credibility of the information in front of them, therefore appears to be vital to further landing credit for EVS/SVS. This will be a key part of flight tests due to be made on the FAA's Boeing 727 testbed in 2005. Fitted with a dual-band Max-Viz EVS-2000 uncooled IR, HGS-4000HUD, MMW radar, prototype HDDs and image fusion processor, the aircraft will also have a modified Rockwell Collins WXR-2100 multi-scan X-band weather radar.
Based on tests of a similarly modified WXR-2100 used in the Reno trials, the company hopes this could be a simple solution to the database integrity dilemma. Rockwell Collins Advanced Technology Center technical director Tim Etherington says: "The weather radar is one possible sensor that we could use as a confirmation of the data. At Reno we had extra digital signal processing software and, in this case, when we lined up on the runway a light came on and confirmed it was clear of obstacles. It formed a real-world connection between the aircraft and the database, and was a huge leap in the comfort factor."
Due to the restricted size of Gulfstream's radome, the antenna on the radar in the Reno test was smaller than Rockwell Collins had planned for the evaluation, which was originally scheduled to use NASA's 757. As a result, the 727 tests will use a full-size antenna and check the fidelity of the modified radar processing software at various angular resolutions and ranges.
Etherington says work on using FLIR and MMW for verification continues in parallel. "We're looking at all sorts of ways to integrate EVS/SVS and bring them together so you can tell which part of the presentation is synthetic and which is enhanced. The simplest is to take the FLIR and put a synthetic wire frame on top, so you can quickly tell if they match. The advantage is you can get a lot more information out of the synthetic picture as you don't have contrast problems and can quickly understand it."
Swift comprehension
Swift comprehension is therefore key to both situational awareness and confidence that what you are seeing is what you are getting. "The sensor [EVS] view picture is going to have a lot more than I really need. For example, as I get closer to the ground I'm going to see houses and other things I don't really need. Or the sensor view can create illusions, such as long shadows in valleys. They can create a picture of things that aren't really there and the integration of synthetic systems can help there," says Etherington.
Another approach is being considered by Kollsman, which is working on a second-generation IR sensor, EVS-II, for FedEx Express - the first airline to equip its fleet with an EVS. Kollsman Avionics product manager Edward Popeck says: "You could look at integrity from the other way around. You could use EVS as the baseline, and an SVS as the verification tool. It should make it easier to certify."
Kollsman is discussing the approach as it works with team member Honeywell on FedEx's $35 million Magic Window EVS effort under which it will equip 200 Airbus A300/A310 and Boeing MD-10/MD-11 freighters from 2007. The carrier's Airbus A380-800Fs are expected to be delivered with HUD/EVS fitted.
NASA SVS project manager Dan Bize believes the key will be developing SVS as a primary navigation system, as the FAA has "established the integrity requirements for navigation purposes. So the challenge is how to meet the navigation requirements once you admit the SVS will be used for primary nav rather than situational awareness. NASA's intent is to make it a flight-critical system, and the industry understands the certification challenge of that approach."
Bize adds that NASA is "working hard" on the integration solution using two possible approaches - one through "an EVS working with an integrity monitor managed by the pilot", and a second "non-human in-the-loop" solution in which integrity will be "verified automatically by a separate computer". NASA is also continuing to develop advanced display concepts for SVS, including differing field-of-view requirements, while in parallel working on a special effort dubbed "high-integrity data", which will be tested in late 2006. NASA's target is to develop it to a technology readiness level (TRL) of 6 (ready to release for potential pre-development) for high-integrity data by 2011.
Business challenge
Bize says NASA "feels like its already achieved" TRL 6 or equivalent in the overall EVS/SVS tests, and companies such as Rockwell Collins believe the focus will soon be shifting from research to applications, particularly if there is the growing expectation of achieving extra landing credit, says Etherington. "There are no more serious technical hurdles. It's a case of finding the best business solution. The biggest problem is the economic climate."
A large part of the process of spreading the word on EVS/SVS is also therefore dependent not only on improving payback to operators, but in cutting the basic costs of the systems. NASA recognises the crucial need for a low-cost approach to developing an integrity monitoring system for a general aviation SVS, and recently flew abi-static GPS array on the Gulfstream V.
Oregon-based Max-Viz is in the vanguard of the low-cost EVS movement, with several applications certificated or in process with its single- and dual-band EVS-1000/2000 IR systems, respectively. "We are seeing increasing demand for EVS products and no doubt that the HUD is an important next step for us," says company chief executive Jim Tuttle. Max-Viz introduced its relatively simple uncooled EVS-1000 sensor as a basic situational-awareness aid optimised for terrain and a follow-on, cooled EVS-2000 dual-band sensor optimised for runway light detection.
Certificated on business jets ranging from the Challenger 601 and Global Express to Dassault Falcon 900, the system is also cleared for use on Bell 212/412 and Sikorsky S-76 helicopters and the Pilatus PC-12 turboprop. "We're on the technology and cost learning curves, and those will get us towards another huge target - the piston singles," says Tuttle, who adds that entry-level and very light jets are also a natural market opportunity.
CMC Electronics of Canada is also examining its future strategy as it prepares to embark on full EVS certification flight tests this month with the SureSight I-series 1-5µ cooled IR sensor on Bombardier's Global Express. Teamed with Thales, which provides a 40¡ field-of-view HUD as well as being system integrator, the CMC SureSight will be the baseline EVS sensor on Bombardier's EFVS, which be installed as standard on the Global Express and offered as an option on the Global 5000.
"We are looking at integrating with SVS and also MMW radar, but there are three issues," says CMC EVS marketing manager Mike Venables. These include the current large size of the antennas required for high-fidelity data, versus the restricted real estate in the nose of the average high-performance business jet, the ever-present price issue and, lastly, the quality of the image itself. "One of the things you want to be able to see is light, but the MMW won't see light. What it will see, however, will be the light-support structures."
CMC is canvassing the industry to determine its next move on a suitable MMW, but "at this point there are no world beaters out there", adds Venables.
Low-cost systems
Higher performance of a level beyond anything yet deployed in the civil world is the goal of a joint Boeing Phantom Works, AFRL Dual Use Science and Technology (Dust) programme to evaluate fused sensor and synthetic imagery on helmet-mounted (HMD), HUD and HDD displays. To be tested in 2005 on a USAF Boeing C-17, the suite includes three FLIR sensors providing a 192¡ wide field of view. Using a BAE Systems head-tracked, visor-projected HMD, the pilot will be able to look anywhere within the fused EVS/SVS image, including down "through" the instrument panel and an aft-looking IR image of the cargo hold. A BAE-developed digital light engine HUD will display images processed at high speeds using field-programmable gate arrays. This will reduce latency in images, which will be displayed as close as is feasible to the ideal "single timeframe" speed of 16.6 milliseconds.
Ironically for a technology that has been mostly advanced by the needs and money of the business aviation world, the Dust effort displays the high-end potential that one day could be available for all users.
GUY NORRIS / LOS ANGELES
Source: Flight International
