One date already has come and passed since the National Transportation Safety Board (NTSB) urged the installation four years ago of cockpit video recorders, and another is fast approaching. At a symposium hosted by the board last week to apply pressure for action on its unrequited recommendations for video recorders, a Federal Aviation Administration (FAA) official conceded that the laborious eight-step process to put the recommendations into effect has not even begun.

To use a Vietnam War metaphor, there is no light at the end of the tunnel. In fact, the regulatory tunnel, as it were, hasn’t even been entered.

Since 2000, the NTSB has called for cockpit image recorders to be installed in transport category aircraft as a necessary complement to cockpit voice and digital flight data recorders (CVR/DFDR) already required in these aircraft. Known by the acronym CIR, for cockpit imagery recorder, the NTSB wanted them to be installed in new aircraft coming off the production line as of Jan. 1, 2003, and CIR retrofits on current aircraft to commence Jan. 1, 2005. Unless done so voluntarily by manufacturers and operators, neither date has any effect without an FAA-decreed requirement.

For smaller turbine powered aircraft engaged in full or part-time carriage of passengers, the NTSB wants crash-protected CIRs installed starting in 2007 for new airplanes and retrofitted in existing airplanes starting 2010. Of an estimated 18,000 airplanes in this class, about half are engaged in purely private use. The NTSB wants CIRs installed in the other half engaged in carrying passengers. Presently, these airplanes are not required to be equipped with either a CVR or DFDR, and the NTSB recommendations basically take the course that CIR equipage is a cost-effective substitute, on the grounds that recording the imagery is better than the present situation where nothing is recorded – confounding accident investigations.

The full set of recommendations lies in bureaucratic limbo.

Carol Carmody, the NTSB member who chaired the hearing, said, “The safety board has investigated more than 100 accidents on aircraft without recorders [e.g., CVRs or DFDRs] since the 2000 recommendations were issued.”

“Frequently, we are hampered [by lack of information] in our investigations, and it took longer to reach a probable cause,” she added.

In the last two months, the safety board has investigated 11 accidents in which the aircraft had no recorders.

Carmody and other board members believe that for airplanes already equipped with CVRs/DFDRs, imagery recording can play a vital role, filling in gaps regarding cockpit displays and pilot actions.

Yet the FAA has not yet acted. Tony Fazio, head of the FAA’s office of rulemaking, outlined an eight-step process every rule undergoes. One step includes a cost-benefit analysis and, Fazio said, “Clearly, we would have to be concerned about the costs and benefits.” With more than 10,000 airplanes subsumed under the NTSB’s recommendations (transport category and smaller turbine powered aircraft), and assuming installation costs anywhere from $1,000 to $10,000 per airplane, the total package could range from $10 million to $100 million.

Moreover, Fazio said it takes anywhere from three to five years to put out a new rule. Thus, even if the rulemaking process were begun today, it likely would not cover two aircraft now in development, the Boeing [BA] 7E7 and the Airbus A380.

Carmody asked, “For your eight-step process, you haven’t gotten to step one yet?”

Fazio: “That’s correct.”

Fazio indicated that items mandated by Congress are the top priority, going to the head of the line for rulemaking as a matter of course. Thus, the “country of final assembly” rule was issued within weeks of receiving the congressional requirement (see ASW, July 12). Meanwhile, the cockpit video recorder recommendations have languished for years.

If and when the FAA does issue a proposed rule for CIRs, pilots’ unions are likely to object strongly, decrying the constant video surveillance as an intrusion on their privacy and the imagery of doomed pilots’ last moments as likely to find its way onto the Internet or evening newscasts – the human penchant for voyeurism and the media’s penchant for sensationalism being what they are.

The pilots’ likely objections could be speed bumps on the path to a video recording requirement, or they could turn out to be a major roadblock. “The cockpit image recorder is not the answer” to missing or uncertain data in accident investigations, declared Capt. John Cox, executive air safety chairman for the Air Line Pilots Association (ALPA). Manufacturers’ objections may also need to be taken (Cont’d on p. 4) into account. Ron Swanda of the General Aviation Manufacturers Association (GAMA) said flatly, “Under no circumstances do we see a CIR substituting for a CVR or DFDR.”

March of technology

The argument over cockpit imagery recorders is playing out against an advancing technology. The cameras and the recording technology are increasing in capability while shrinking in terms of size and weight, while at the same time reducing in cost.

Frank Doran of L3 Communications, a manufacturer of cockpit voice and flight data recorders, said a crashworthy camera with a two-hour recording capability can be had for about $10,000. The cost of a total cockpit imagery recording capability depends primarily on the number of cameras. The NTSB wants to obscure the view of the pilots’ heads, restricting the imagery to their hand and arm movements on the controls, and to capture instrument displays. That requirement makes the task difficult to accomplish with a single camera. Doran said the industry could accelerate production of a suitable suite of cameras and related equipment within 18-20 months of a “go” order.

Mike Horne of UK-based AD Aerospace, a camera supplier whose equipment already is installed for cabin surveillance on JetBlue Airways [JBLU], said a single high-resolution camera costs around $5,000-$7,000. However, a low-resolution camera adequate for the NTSB’s desire for imagery recording in general aviation aircraft can be had for around $1,500.

Jim Elliott of Smiths Aerospace, another recorder supplier, said for a transport category aircraft a suite of four cameras would provide desired coverage of the cockpit as well as of the cabin: one camera covering the left side of the cockpit, one camera covering the right side of the cockpit, one camera looking over the pilots shoulders to cover the center console, and one camera to cover the cabin.

Rick Shie said his small ($22 million annual revenue) California-based company, Physical Optics Corp., is developing a single-box system to perform multiple recording missions. Dubbed FAERITO(tm), for flight anomalous event recorder information technology, its development is being supported by the U.S. Navy under a small business innovative research (SBIR) contract for installation in the Navy’s E2C Hawkeye aircraft. The Navy has a requirement to install a recorder in these airplanes, which presently are not so equipped.

The FAERITO incorporates separate recorders in one survivable steel box, to include video, voice, data and engine recorders. The box itself measures 4″ x 5″ x 7″ and weights about nine pounds.

FAERITO is being developed with an eye for civilian sales. The company hopes to receive FAA certification by April 2005, and it will be available for commercial application in July 2005, Shie said.

The power of precedent

Video is finding increasing application in aviation training. For example, Navy T-45 training aircraft feature a camera recording the head up display (HUD). The recording is used to debrief student pilots after their training flights.

In addition, the Navy has developed a prototype system in which video is used to record simulator training. The video is used to debrief the pilots after the training session, to point out what they did well and to highlight errors of omission or commission. “We have the entire evolution of the training event for the debrief,” said Constance Gillan of the Navy’s Pacific Fleet Sea Control Wing. She has been part of the development of this technology for the past seven years. “Video combined with other sources is the way to go,” she declared. Gillan hastened to add that this is not her position as developer, but the reaction of pilots, who have submitted reports, such as one student pilot who praised the video as the “best simulator improvement per dollars spent.”

Gillan said the prototype system has been so successful that it is being installed on simulators in operational use.

Although cameras are being exploited by the Navy primarily for enhanced training and not necessarily for safety, the fact that the video recordings can be replayed, scrolled back to precursor events, or scrolled forward to highlight consequences, points to the obvious application of such a capability in accident investigations.

‘Missing link’ vs. ‘misleading’

ALPA’s Cox maintained that video playbacks in accident investigations could be misleading. “The audio data has more clarity,” he said. “My concern centers on the subjectivity of the images, which could mislead the course of an investigation,” Cox declared.

He cited as an example the 1994 crash of USAir Flight 427, which the NTSB ultimately concluded was due to an uncommanded rudder movement. If a CIR had been installed, Cox said, investigators might have had difficulty determining whether the pilot was pushing on the rudder pedal or the pedal was pushing on the pilot. “A force transducer would answer that question,” Cox said, by way of suggesting that improved DFDRs would provide more insight for investigators than a CIR.

Ken Smart, chief inspector of air accidents for the UK’s Air Accidents Investigation Branch (AAIB), noted that his agency has called for CIRs as a result of challenges it has faced reconstructing the chain of events in some high-profile accidents.

“The DFDR and CVR are not giving a good representation of what the crew is trying to confront,” Smart said, especially in those cases involving “a lot of dramatic events going on at the same time.”

“It is common on the CVR to hear words like ‘look at this’ and not being able to determine what ‘this’ is,” Smart recalled.

As another example, Smart said the issue of which pilot is flying the airplane “comes up time after time, especially when things start to go wrong.” He said. A CIR would aid in deciphering such conundrums, he said.

Smart was one of a number of witnesses who pointed out the degree to which human factors issues play in the majority of accident investigations. “If we are to understand more fully the human factors issues, a CIR can help,” he said. Imagery of the cockpit, Smart said, is the “missing link” in accident investigations, and a much-needed complement to the information gleaned from CVRs and DFDRs.

Indeed, the Navy has found that many human factors issues come to the fore in its simulator training – issues all-too-often involved in greater or lesser degree in accident postmortems.

Going global

The NTSB and the AAIB are not the only organizations that have called for cockpit imagery recorders. The most recent development comes from France. On July 7, Madame Odile Saugues, president of La Mission d’Information Sur la Securite du Transport Aerien de Voyageurs, submitted a report to the French National Assembly on a number of initiatives, one of which included a call for cockpit video. The recommend-ation relates to the Jan. 3 crash of an Egyptian Flash Air B737 on climb out of Sharm el-Sheikh, in which all 148 passengers (mostly French tourists returning to France from holiday) were killed. From the report:

“The installation of a video camera inside the cockpit would indeed prove particularly useful in the case of an air disaster. In the dramatic case of Sharm el-Sheikh, based on the information released to date in the media, the catastrophe may have been caused by a glitch in the functioning of the automatic pilot of the Boeing B737. The recording of the acoustic environment inside the cockpit seems to indicate that the automatic pilot had been turned on. On this issue – which is crucial to determine the causes of the accident – the installation of a video camera would have added some information extremely useful to the investigation.”

“Proposal: Anticipate the mandatory installation scheduled by ICAO [International Civil Aviation Organization] on Jan. 1, 2007, of fast access flight recorders in aircraft over 10 tons, and introduce, in coordination with the builders, the mandatory installation inside the cockpit of a third black box which could not be tampered with, sheltering a video camera to film the control panel. The video recordings would be analyzed only in case of an accident, excluding any other use by the airlines.” Source: http://www.assemblee-nationale.fr/12/pdf/rap-info/i1717.pdf, p. 234

The basic technical specifications have already been written. In March 2003 the European Organization for Civil Aviation Equipment (EUROCAE) produced a minimum operational performance specification (MOPS) for image recording in the cockpit (see ASW, June 9, 2003). Known as EUROCAE document ED-112, it does not spell out the number of cameras to install, but rather sets forth the desired performance in terms of update rate (4-5 frames per second), field of view (general cockpit area) and resolution (sufficient to read the instrument panel). This document, witnesses at the NTSB hearing said, could serve as the basis for an FAA technical standard order (TSO), a necessary adjunct to the rulemaking process.

Feeding frenzy or faster facts

ALPA’s Cox questioned whether CIRs are needed at all. With no unsolved accidents involving air carrier aircraft in the last 20 years, he questioned the need for CIRs. With respect to that history, Carmody replied, “Frequently, we were hampered in our investigations, and it took us longer to reach probable cause.”

Pilots union representatives expressed enormous concern that cockpit video recordings of a crash would leak to the public. The case of the American Airlines [AMR] 1995 crash into a ridgeline at Cali, Columbia, stands as a case in point. The Colombian government released the tape of the cockpit voice recorder, which received wide play in the media.

James Johnson, an ALPA lawyer, said, “Cali is a warning that we don’t have adequate protections outside the United States, and we need to look at that.”

The NTSB’s Chris Julius said, “We need to see if we can balance the privacy concerns with the investigative interests.”

Johnson countered, “To us, there isn’t a balance and privacy overrules.”

“We don’t object to the use of the CVR for accident investigative purposes, but we have pilots around the world being prosecuted for operational errors, so if we go the CIR route, the protections need to be strengthened,” Johnson insisted.

Lawyers would have a “feeding frenzy,” Johnson envisioned. A picture is worth a thousand words, and the video would be used by plaintiffs lawyers to dramatize pain and suffering, he theorized.

Erasing the CIR tape after the NTSB completes its investigation was generally thought to be a non-starter. “I don’t think this is a good idea,” declared Michael Demetrio, a plaintiffs lawyer from Chicago. CVR, DFDR and CIR materials are all part of the record and should be open to analysis, and certainly available under discovery with appropriate protections against unauthorized use, he argued.

“A CIR would help dismiss silly claims, like the one that the USAir Flight 427 pilot was reading a newspaper,” Demetrio added.

Mark Dombroff, a Washington, D.C.-based lawyer with airline clients, said, “If the CIR is the only recorder, release under protective orders to lawyers and use in trials is even more likely.”

Controlling access to imagery may be particularly difficult in international cases. The AAIB’s Smart outlined a three-key concept by which investigators could gain access to encrypted imagery. Authorities of the country conducting the accident investigation would have one key, the CIR manufacturer would have another, and representatives of the pilots’ union would have the third. All three keys would be required to gain access.

Research results

Further impetus for CIRs may come from ongoing research slated for completion in September. Conducted by Pippa Moore of the UK’s Civil Aviation Authority (CAA), the research was undertaken to compare the data provided by a CIR against the data provided by CVRs and DFDRs and whether “any additional information is provided and whether the benefits associated with the additional information justify the potential invasion of flight crew privacy,” according to the latest (June 18) update on this research.

The interim conclusions affirm that:

  • It is possible to install imaging recording systems that provide a general view of the flight deck without exposing the identity of the flight crew.
  • The image recordings provide useful data not available for the CVR or DFDR (e.g., loss of flight displays and failed flight crew attempts to solve problems).
  • Image recorders can confirm facts suggested by other recorders (e.g., smoke in the flight deck).

AD Aerospace’s Mike Horne participated in this CAA simulator research. He recalled that pilots tended to point to partial panel failures, for which there was no indication on the CVR or DFDR. In a smoke trial, one pilot knocked off his glasses while donning the smoke mask and goggles, a momentary distraction not caught by the CVR.

Although Horne has an interest in cockpit imagery as a potential camera supplier, he pointed to the NTSB’s challenge regarding FAA inaction to date. “An accident investigation tool will not be installed in airliners without a requirement,” Horne said.

“We need to push,” he declared. “New aircraft are in design now, and now’s a chance,” he added. Regarding the pilots’ privacy concerns, Horne said, “Pilots are fully competent professionals. They’re going to be following procedures, so I really don’t see what the problem is.”

Overall, it was plainly evident during the course of the hearings that the pilots object, suppliers are ready to refine and deploy their designs, and safety board recommendations to install imagery recorders depend utterly on FAA action. >> Smart, e-mail [email protected]; Horne, e-mail [email protected]; Shie, e-mail [email protected]; Gillan, e-mail [email protected]; Demetrio, e-mail [email protected] <<

Call for Cockpit Video NTSB recommendations for cockpit video recorders and present status
Issue Date & Recommendation No. Recommendation Present Status
For transport category aircraft (in passenger/cargo service)
April 2000 #A-00-030 Deals with retrofit Retrofit by Jan. 1, 2005, cockpit video recorders on all aircraft currently required to be equipped with a cockpit voice recorder (CVR) and digital flight data recorder (DFDR). Open – Unacceptable Response
April 2000 #A-00-031 Deals with new aircraft Require installation of cockpit video recorders on all aircraft manufactured after Jan. 1, 2003, currently required to be outfitted with a CVR and DFDR. Open – Unacceptable Response
For all smaller turbine powered aircraft (used full or part-time for commercial or corporate purposes)
December 2003 #A-03-064 Deals with new aircraft Require installation of a crash-protected image recording system on aircraft manufactured after Jan. 1, 2007, not equipped with a CVR or DFDR. Open – Await Response
December 2003 #A-03-065 Deals with retrofit Retrofit by Jan. 1, 2010, a crash-protected image recording system on aircraft not equipped with a CVR or DFDR. Open – Await Response
Source: NTSB