Expert Systems and Collision Regulations - Essay Example

Expert Systems and the Collision Regulations of Presenter] Module of the A surge in intercontinental commerce, population explosion, faster communication and commuting, the need for speed and competition has led to a massive explosion in the number of aircrafts and ships that need to enable transportation of mankind. Traffic congestion is not just a matter of the roads anymore, but also of the sea and the air. Pilots and navigators have to be more alert to intrusions in to their course vector to ensure that no high-level threat situation emerges in their charted course. One to one manual monitoring with electronic radar has become so ineffective with so many targets identified by the gadgets - especially for aircrafts on approach and take off and for ships that call on and off the port and service lanes. Port corridors have become so congested that it is even tough to row through let alone guide a huge vessel to its berth. The upswing in the accident curve has made scientists sit up and think about safer modes to navigation. Application of technology has certainly made it possible for us to have a safer travelling and transport environment. This paper examines the various devices that have been devised to assist navigators and empower them to make knowledgeable decisions in the event of a crisis. Advance warning is a key area in which these radar based electronic devices have managed to improvise leaving the human element to play a vital role to decide on the final course of avoidance action. Introduction The International and Inland Maritime Navigation Rules were formulated in the Convention on International Regulations for Preventing Collisions at Sea Treaty in 1972 and became effective on the 15th of July, 1977. The rules were intended to codify the standard behaviour of vessels of all the nations to substantially reduce the possibility of mid-sea collision by promoting orderly and predictable responses to a variety of frequently occurring situations. The regulations are as under the following categories: Conduct / Manoeuvre of ships in any condition of visibility Priority and precedence of various types of vessels Nighttime lighting of vessels Daytime day shapes of vessels Sound signals for communication of actions, intents and status of ships In the case of aircrafts, they too require such laws and international standards. In the initial days of flying, the pilots trusted their eyes and ears to do the job. All through World War I and later mail aviation happened without any sophisticated navigation equipment except for a front mounted compass. Later as technology developed, wireless communication systems were installed and during the Second World War, an airborne radar was fitted to the aircrafts -not initially to assist navigators to fly, but to hunt down submarines and ships and to shoot at them. Later, as civil aviation grew in volume and planes were taking off airstrips more often than earlier a combination of these two systems - the wireless and the radar helped pilot make decisions that avoided mid air collisions. These systems had their risk. If the pilot was not experienced enough, he could not make the knowledge calculations to prevent a collision in a crisis situation. Huge efforts were being made to train pilots to retain their sanity at the time of a crisis. The introduction of the jet age compounded the situation. Aircrafts were now flying higher and faster. The pilots had less response times at hand. Research showed that the aircraft would easily travel a mile along its horizontal displacement vector between the time the pilot pulled the stick and the aircraft even begins to climb. This is a huge gap for error. Needle line consciousness of pilots prevented accidents from occurring. But more planes were lost due to mid air collisions than at landing and take off - most of them quite unjustly described as pilot errors. The complications of aircrafts were constraining to both the regulators and the inventors of the system. Aircrafts could not carry heavy equipment to control their weight. Hence any equipment that the plane carried had to be lighter than the ones fitted on the ships. The second reason was even more challenging. The equipment had to fit in to the existing cockpit of the pilot that was already crammed and running out of space. For two decades no compromise formulas could be arrived at. Research in to developing an advance collision warning system had already began in 1950s. What led the authorities to swing in to action were the Grand Canyon collision of two airliners and the PSA - Cessna collision in 1978. The congressional legislation mandating Traffic Alert / Collision Avoidance System was inspired by the 1986 collision of Aeromexico DC 9 and a private US plane on August 31st. Throughout this period, many versions of avoidance systems and methods were proposed. But they were doggedly denied by the pilots and aircraft companies as inefficient or not suitable for flying machines. TCAS or other similar systems had been through various levels of development since the mid 50's. Several of them had been developed and used by the maritime community where limitations posed by space and weight constrains were much easier to negotiate. Research during this period confirmed that most collisions occurred while flights overtook each other. Research had also confirmed the horrific fact that a mere warning to the pilot was not enough to prevent a mid-air collision. Hence to serve any practical purpose, the pilot should be supplied with the relative bearing information that a threat proposes and in so doing, he has to be given enough time in advance to negotiate a method to eliminate or reduce the threat. With all these facts in mind, scientists all over the world sat down to work to develop a new piece of equipment to be installed to aircrafts in particular to protect it from midair collisions. The research was know as the Airborne Collision Avoidance System Research or ACAS. One of the earliest system proposed was a three-range device for high-speed jet aircrafts. It was based on an adjustable system that would lower the incidence of false alarms. The short range was to be used in airport corridors and congested near terminal environment. The medium range for low altitude cruising and the long range fro high altitude flights. The earliest systems were based on an attempt to calculate the miss distance or the proximity to the target at which the system would raise an alert. For the system to work accurately, it needed to trace the relative bearing angle and closing rate to be calculated on a realtime basis. Turbulent conditions were a challenge for the system and the instruments proposed and developed were also bulky. The system had no buyers. Another type system, developed by Bendix Radio in the late '50s and early '60s, used a different approach. It used to calculate the time left for the participating aircrafts to still evolve a safe manoeuvre. The system was a developmental success because it did not attempt to predict the miss distance. Therefore it could be used in turbulent conditions as well. Before reaching the minimum separation, it would advise the pilot to climb or dive. The vertical component of the instrument operated with a small UHF transmitter which periodically transmits a series of pulses. The pulses were spaced at varying intervals based on the aircraft's altitude. The receiver in the system checked for any similarly pulsing aircraft in the vicinity. More analysis was required if an aircraft was detected at or near the same altitude. It was during the development of the Bendix system, that Dr. John "Smiley" Morrell discovered the concept of "Tau". Tau is based on time and not distance. In mathematics, Tau is expressed as the range to the intruder divided by its closure rate or range-rate. The Bendix system needed to determine the range of the aircraft at or near the same altitude and the speed at which the range varied. Engineers accomplished this fete by devising a system called the ground-bounce ranging system. A transmitter sent a split signal - one that travels directly to the receiver and another that bounced off of the ground, to the aircraft. The delay between the direct signal and the ground-reflected signal was calculated to further determine how far apart the aircrafts were. If the delay was short, the aircraft were separated considerably, and if the delay was long, the target aircraft was within close proximity. If the ratio of range to range-rate reached twice the minimum escape time for the type aircraft on which the system was installed, the alarm sounded. took a different approach and used time to determine how long before participating aircraft obtained their separation minimum with there still being enough time to escape. This system was deemed more efficient because it did not try to predict "miss-distance", therefore the problem of accurate bearing measurement which plagued the previous model was not a factor. Before reaching minimum separation and in enough time to evade the intruder, an alarm would sound and tell the pilot to climb or dive. The vertical component of the system operated with a small UHF transmitter which periodically transmits a series of pulses. The pulses were spaced at different intervals based on the aircraft's altitude. The receiver in the system interpreted the altitude of any similarly equipped aircraft in the vicinity. Further analysis was required if an aircraft was detected at or near the same altitude.5 During the development of the Bendix system, Dr. John "Smiley" Morrell discovered and first used the concept of "Tau". Tau is based on time, not distance. Mathematically, Tau is expressed as the range to the intruder divided by its closure rate or range-rate.6 The Bendix system only needed to determine the range of the aircraft at or near the same altitude and the rate at which the range changed. Engineers did this by devising a system called the ground-bounce ranging system. A transmitter sent a split signal one that traveled directly to the receiver and another that bounced off of the ground, then to the aircraft. The time delay between the direct signal and the ground-reflected signal was calculated to determine how far apart the aircraft were. If the delay was short, the aircraft were separated considerably, and if the delay was long, the aircraft were within close proximity of each other. If the ratio of range to range-rate reached twice the minimum escape time for the type aircraft on which the system was installed, the alarm sounded. The Eliminate Range-zero System (EROS) was designed for fast moving, fighter-type aircrafts. EROS put to use time-frequency techniques. Aircrafts carried an accurate cesium-rubidium clock that was synchronised to a master clock. A pulse train of information (including the host airplane's altitude) is transmitted at a time precisely allocated for that airplane. Based on the time differential calculated when another airplane received the signal, EROS could determine the range and closing speed of the approaching airplane. The drawback with this system, like all of the others that preceded it, was that it only protected against aircraft with the same equipment on board. Moreover, this system was prohibitively expensive as well. Hence it was considered impractical and was never used. Since the mid '70s, efforts have been focussed on value addition of hardware already installed on most aircraft, namely the transponder of the Air Traffic Control Radar Beacon System (ATCRBS). Basically all aircrafts would be installed with airborne interrogators that would be able to interpret data from the transponders of nearby aircraft on board. Such systems became to be known as the Beacon Collision Avoidance System or BCAS. In the late '70s, George Litchford, a New York electronics engineer, came up with a theory that a passive anti-collision system could eavesdrop on ground interrogators that search and locate and track nearby aircraft. It was given the name passive BCAS. This technique is based on listening for transponder replies from nearby aircraft in the vicinity to two or more ground interrogators. By timing the receipt of these ground interrogations and replies from another aircraft, and using the known positions of the ground interrogators, a passive system worked out the relative positions and altitudes of other aircraft. Passive BCAS never went into full production because it was considered too complex and would not work over the ocean or in places where there was limited or no radar coverage. However, with the electronic and navigational capabilities that exist today, there is ample scope for a passive TCAS system. This is because in some instances, aircraft know their coordinates exactly if they are navigating with INS, Loran-C, or GPS. A passive system would only need to receive a signal from one ground interrogator. TCAS is a relatively simple system that works quite effectively. In very simple terms, the system identifies the location and tracks the progress of aircraft equipped with beacon transponders. Currently, there are three versions of the TCAS system that are in use or in some stages of development. They are the TCAS I, II, and III. TCAS I, the simplest of the systems, is less expensive but also less effective than the others. It was designed primarily for general aviation use. The TCAS I transmitter sends signals and interrogates for Mode-C transponders in the vicinity. The TCAS I receiver and display shows the pilot approximate bearing and relative altitude of all aircraft within the selected range, usually about forty miles. Further, the system uses colour-coded dots to indicate which aircraft in the area pose a potential threat. This is referred to as a Traffic Advisory (TA). When a pilot receives a TA, it is up to him to manually identify the intruder and decide on the course of action; he is permitted to deviate up to + 300 feet. Lateral deviation however is un-authorised. In instrument conditions, the pilot is required to notify the air traffic controller for his cooperation in resolving the conflict. TCAS II on the other hand is more complicated and comprehensive system than TCAS I. This system was mandatory on all commercial air carriers operating in the United States by December 31, 1993. It offers all of the same benefits while also issuing a Resolution Advisory (RA) to the pilot. In other words, the intruder trajectory is plotted and the system is able to tell whether the aircraft if climbing, diving, or in straight and level flight. Once this is determined, the system will advise the pilot to execute an evasive maneuver that will take away the aircraft from the intruder. There are two types of RAs, namely preventive and positive. While preventive RAs instruct the pilot not to change altitude or heading to avoid a potential conflict, positive RAs indicate to the pilot to climb or descend at a predetermined rate of 2500 feet per minute to avoid a conflict situation. TCAS II is capable of interrogating both Mode-C and Mode-S. In the case of both aircraft having Mode-S interrogation capability, the TCAS II systems communicate with each other and issue resolved RAs. Since this system costs upward of $200,000 per aircraft, manufacturers have accommodated an upgrade option as well to the next generation TCAS III. This system will be virtually the same as TCAS II but will also allow pilots who receive such RAs to execute lateral deviations to evade targets. This will be possible because by the use of a directional antenna on TCAS III that will be more accurate and will have a smaller bearing error. The new antenna will cut down on false alarms since it can more accurately determine an intruder's location. Another proposed upgrade is the incorporation of the Mode-S data link. With this link, the system will be capable of transmitting the aircraft's GPS position and velocity vector to the other TCAS-equipped aircraft thus providing much more accurate information to avoid or resolve crisis situation. REFERENCES Riley, Gary, CLIPS Reference Manual Vol. 1: Basic Programming Guide, Version 6.1, August 1998. United States Coast Guard, Navigation Rules, International - Inland, COMDTINST M16672.2D. U.S. Government Printing Office, March 1999. Stottler, R., and Vinkavich, M., "Tactical Action Officer Intelligent Tutoring System," Industry/Interservice Training, Simulation, and Education Conference (IITSEC 2000), Orlando, FL, November 2000. Lee, J., and Sanquist, T., "A Systematic Evaluation of Technological Innovation: A Case Study of Ship Navigation," Proceedings of the 1993 IEEE International Conference on Systems, Man, and Cybernetics, Vol. 1, Le Touquet, France, 1993, pp. 102-107. Brown, J.S., and Sleeman, D., Intelligent Tutoring Systems. Academic Press, Inc., 1982. Rowe, N. and Galvin, T., "An Authoring System for Intelligent Procedural-Skill Tutors," IEEE Intelligent Systems, Vol. 13, No. 3 (May/June 1998), pp. 61-69. Rowe, N., and Schiavo, S., "An Intelligent Tutor for Intrusion Detection on Computer Systems," Computers and Education, Vol. 31 (1998), pp. 395-404. Klass, Philip J., "New TCAS Software Cuts Conflict Alerts", Aviation Week & Space Technology, September 20, 1993, p. 44. McClellan, J. Mac, "Collision Vision", Flying, May 1989, p. 54-56. Reingold, Lester, "TCAS: Not-Quite-Perfect Solution", Air Transport World, January 1992, pp. 78-80. Westlake, Michael, "How to Avoid Air Collisions", Far Eastern Economic Review, December 20, 1990, p. 66. "FAA Redirects TCAS-III Effort", Aviation Week & Space Technology, September 27, 1993, p. 37. "United to Test TCAS Use for Altitude Changes", Aviation Week & Space Technology, November 22, 1993, p. 63. Read More