In order to reduce the cost and improve the safety of travel to and from orbit sufficiently to enable space activities to become routine and commercially profitable, fully reusable launch vehicles operated like aircraft are essential. Aircraft are built, tested, and operated in a manner that achieves extremely high levels of safety, through the implementation of a safety-oriented operating philosophy that has been developed and improved progressively over decades of operation throughout the world.

A newly-built aircraft - even the 1000th example of a long-established line - is subjected to a series of flight tests which progressively expand the operating envelope. This done, the vehicle is placed in operational service. Every component of the vehicle has a specified Time Between Overhauls (TBO) based on extensive and rigorous testing. After each flight the number of flight hours on the different components are compared to their TBO, and, if necessary, components are removed and replaced with spares. Likewise, of course, any failed systems are changed out. Extensive changes (non-routine change-out of an engine, major rewiring, etc) will result in a Functional Check Flight - effectively another test flight to ensure that the newly partially-rebuilt aircraft works properly.

Elaborate inspections, clean-room efficiency, and the swarms of inspectors commonly involved in the launch of an Expendable Launch Vehicle ( ELV) are not necessary before each flight of an aircraft for two main reasons. First, the consequences of a failure are survivable: if an engine, generator or other component fails, the aircraft simply lands. Secondly, the sure knowledge that a component worked on the last flight, as on hundreds of previous flights, provides greater assurance of reliability than is provided by removal, inspection and re-installation, which introduce new possibilities of installation error and component failure.

In contrast to aircraft, ELVs have no emergency procedures after launch except to destroy the vehicle with explosive charges! NASA's partly re-usable Space Shuttle does have emergency procedures, but they are applicable only in very narrow segments of the ascent flight regime, and are of marginal utility even then. There is no operational equivalent of an aircraft aborting down the runway and taxiing back to resolve a maintenance fault. Nor of landing immediately after takeoff: the Space Shuttle's "Return To Launch Site" abort is problematic at best, while the aborts downrange to locations such as Dakar are available only for a short period, and possible only for some malfunctions. In addition, such an abort is just as enormously expensive as a nominal Space Shuttle launch - hundreds of $millions, which is some 10,000 times more than the tens of $thousands cost per flight typical of a large airliner. Effectively, once it is committed to launch, a Space Shuttle is not savable. For this reason, its operation resembles ELV operations much more closely than aircraft operations.

Because there is no way to abort the flight of an ELV once launched, it is necessary to ensure first-flight reliability through extensive ground testing. This method works moderately well, but at great expense. In adopting ELV-like procedures, the Space Shuttle acquires the same disadvantages, and throws away the great reliability advantage of reusable vehicles - the knowledge that an installed system which worked on every previous flight can reasonably be expected to work the next flight, barring accident or exceeding TBO. Instead, after a Space Shuttle flight, working systems are dismounted, inspected, refurbished, and extensively tested before being re-mounted (usually on a different shuttle on a later flight).

ELV's and the Space Shuttle are thus vehicles which must successfully complete their first test flight of an effectively brand-new, very complicated vehicle by flawlessly performing a very demanding flight. Failure of nearly any component results in loss of the vehicle and, if applicable, crew. The fact that this method works at all is a testament to the skill and effectiveness of the people and tests involved. However, there is a better way.

Airliners achieve far greater reliability than launch vehicles, at far lower cost. This has nothing to do with the technology involved, nor the type of flight. Rocket engines are in some respects stressed more severely than jet engines - but not greatly so. Flight to and from space is more punishing than high-altitude and high-speed aircraft flight - but not greatly so. The difference in reliability has everything to do with the method of operation. A properly designed and operated space vehicle could achieve the same reliability as an aircraft - but this requires the adoption of "aviation philosophy" towards safety and cost-effectiveness in both design and operations.

A very important feature of "aviation philosophy" is that airplanes are piloted - as indeed are all passenger-carrying vehicles, except for a small number of short-distance "people movers". Based on the enormous experience of the aviation industry accumulated over hundreds of millions of flights carried out over almost a century throughout the world, piloted vehicles are considered to be fundamentally safer than unpiloted vehicles. The possible use of unpiloted aircraft for passenger transportation does not receive any serious attention, even though it could potentially reduce costs.

By contrast, all launches to date have involved expendable rockets, and piloted launches are considered more dangerous than unpiloted launches due to the increased risk of loss of life. However, as discussed above, expendable rockets are essentially missiles, and their operation is unavoidably much less reliable than aircraft.

In this paper it is assumed as fundamental that fully-reusable, passenger-carrying rockets traveling to and from orbit will be piloted, and that their operating procedures will be much closer to existing aviation procedures than to the procedures used for launch vehicles used to date.

SPECIAL FEATURES OF VTOVL ROCKETS

The operation of a VTOVL rocket such as Kankoh-maru will be significantly different from operating an aircraft in a number of ways, which raise a range of interesting new issues for pilots and for vehicle designers. The most significant difference is that Kankoh-Maru has no wings, and therefore is not supported by lift; it is supported solely by the thrust of its rocket engines. This will make Kankoh-maru's operational reliability particularly dependent on engine reliability, and in many respects its operations will resemble those of helicopters or the McDonnell Douglas DC-X VTOL test vehicle more than airplanes. In the following sub-sections some of these operational differences are discussed.

Kankoh-maru's thrust-to-weight ratio will initially be 1.4 and will rise to 3 or more, which is some 5-12 times higher than the roughly 0.25 of airliners and cargo planes (though only 2-3 times higher than for high-performance fighter aircraft). This has implications for Kankoh-maru's flight stability, since reactions to changes in thrust from the propulsion system will be relatively faster than that of most aircraft. Consequently the reaction-time required of Kankoh-maru's pilot would be shorter than for commercial aircraft. In general, the pilot of an airliner could, in principle, control the vehicle without the help of the on-board computer. In the case of Kankoh-maru the reaction-time required of pilots has yet to be determined experimentally. Fighter aircraft require the pilot to react faster than commercial aircraft, and some modern fighter planes are inherently unstable, and are reputedly uncontrollable in the event of a computer failure. For Kankoh-maru the relation between the pilot and the control system may therefore be closer to that in a fighter plane than in an airliner or transport plane.

The details of Kankoh-maru's propulsion control system will depend on such details as the response-time of the engines to changes in throttle-settings, the layout of the engines in the vehicle, and the detailed design of the engine control system, down to the level of individual engines. For example, for some operations an automatic link may be made between opposing pairs of engines in order to minimize any thrust imbalance in the event of anomalies. It should be noted that while Kankoh-maru's stability under thrust is presently indeterminate, an advantage of its base-first design is that it is always stable during unpowered descent, due to the low center of mass. Stability in powered descent and hover with largely empty propellant tanks, as after re-entry, seems unlikely to be problematic due to the low thrust involved, although this remains to be verified by simulation and experiment. Thus a failure that results in relying on the pilot for control to landing will generally result in a safe abort. The outstanding questions requiring investigation concern stability when climbing.

In this context it should be noted that the requirements for commercial passenger transport and for cargo transport are somewhat different. Most flights of modern commercial airliners are routine flights between large airports, repeated many times. Cockpit automation has progressed so far that for most phases of a flight, from take-off to landing, active inputs by the flight crew are unnecessary, although of course they retain responsibility for the flight. By contrast, cargo vehicles are likely to be used for more non-routine flight operations for which, although on-board computers will be used, the flight crew will be more actively involved than in scheduled passenger flights.

A second difference between Kankoh-maru and aircraft is that Kankoh-maru's flight-stability is maintained by both aerodynamic controls and by reaction-controls at different phases of its flight. Under normal conditions the pilot of a routine commercial Kankoh-maru flight need not even be aware of which system is being used at any particular time - this will be decided "seamlessly" by the computerized flight-control system. However, in order to be able to take appropriate emergency action, the flight crew must know which system is in operation, and must have practice of operating both systems under emergency conditions.

In the level of automation, the role of Kankoh-maru's pilot on a commercial passenger flight may not be very different from that of the pilot of a modern jet airliner with a largely automated "fly-by-wire" control system. There will nevertheless be significant differences in the details of the control system, and specifically in the scope of pilot authority within the vehicle's control system under different fault conditions.

A third difference is in Kankoh-maru's flight path and its integration with existing air traffic management systems. When departing, Kankoh-maru will take off vertically, and will climb almost vertically though usually towards the east, until it is out of the atmosphere. When returning from orbit, Kankoh-maru will generally fly back from re-entry traveling in an easterly direction, though often coming from either a southerly or northerly direction, and will land vertically. For sub-orbital flights also, Kankoh-maru's take-off and landing trajectories will be almost vertical, although its direction at high altitude will be more varied. It is intended that Kankoh-maru will operate from commercial airports (2), but even so the time spent in controlled airspace will be very short, and so it should not be difficult to integrate these routes with existing flight corridors. It may be considered desirable for new vertical departure and arrival routes to be formally decided for airports handling Kankoh-maru and other VTOVL vehicles, analogous to the SIDs and STARs used by aircraft (See Glossary of Terms below).

Fourth, it will not be economic for Kankoh-maru to carry significant amounts of spare propellant on scheduled flights to and from orbit. Consequently Kankoh-maru will have short launch windows, since it will not carry sufficient propellant to be able to make substantial orbital plane changes. And it will also have very short landing windows, since it will not be able to hover for any significant length of time before landing. Kankoh-maru will therefore need to be given priority over other aircraft for the short periods of time during which it is using controlled airspace.

In this respect a normal Kankoh-maru landing will have features similar to the procedure for a "Lifeguard Flight" or an emergency landing by an aircraft. It will however be less inconvenient for other aircraft for several reasons: a Kankoh-maru landing will be scheduled hours in advance and so will involve no last-minute change of flight plans; its impact on other air traffic will last for no more than a few minutes before and after take-off, or from atmospheric re-entry to landing; Kankoh-maru will not use an aircraft runway, but a separate and much smaller landing site; emergency services will not need to be called to the landing-site; and finally, only a rather small number of airports will be involved. This is because a traffic rate of even 1 million passengers per year will represent only some 60 flights per day world-wide, which could be handled by quite a small number of airports.

Under normal circumstances, in which scheduled airline flights are following their planned schedules closely, other aircraft would not be inconvenienced, nor even aware of the priority being given to a Kankoh-maru taking-off or landing. It is also conceivable that, once Kankoh-maru operations become routine, they may be scheduled in parallel with other aircraft landing and taking-off, as some airports currently operate two parallel runways simultaneously.

In the improbable case where another aircraft had to make an emergency landing at the same time and at the same airport as a Kankoh-maru was landing, giving priority to Kankoh-maru would be unlikely to have a critical influence since it would last such a short time. In such an emergency Kankoh-maru could, if necessary, even land outside the airport - but it could not not land. Overall, giving Kankoh-maru airspace priority would have less impact than the special provisions agreed for Concorde's supersonic flights, since the duration of its atmospheric flight will be a matter of only minutes rather than the hours for which Concorde flights last. (Other airliners are not permitted to fly within a specified corridor directly below Concordes which are flying supersonically, since they may descend rapidly without warning in the event of a solar flare causing raised radiation levels in the passenger cabin.)

This Flight Manual provides an outline of the detailed operating procedures for a Kankoh-maru version designed for operation as an all-purpose transport vehicle for both orbital and sub-orbital flights. It includes the division of tasks between the various crew members. Particular emphasis is given to the emergency procedures, which are critical for achieving the high level of safety required for routine passenger-carrying operation. A number of assumptions have been made about the vehicle's sub-systems and their operation; the precise details will depend on the vehicle's final design.

The Flight Manual is also intended to give a true-to-life feeling of Kankoh-maru's aircraft-like operations. It is hoped that this will show that, whereas existing launch vehicle operations require "swarms of PhDs in lab-coats", Kankoh-maru can be operated like any other airplane - as the USAF likes to say, "maintained by high school graduates and flown by history majors."

Accordingly, this Flight Manual reads like the "Dash One" manual of an operational aircraft. The document is modeled after the "Dash One" used in USAF aircraft since that is the type of document with which one of the authors is most familiar. US Navy NATOPS, foreign military manuals, and airline operating manuals provide essentially the same guidance, though not always in the same format. For purposes of this hypothetical "Dash One" extract, it is assumed that three different examples of Kankoh-maru have been acquired by the USAF to serve as all-purpose orbital and surface-to-surface ballistic transport vehicles. Features incorporated in the Kankoh-Maru for dedicated passenger transport are retained, with the exception that it is assumed that the seats themselves are removable to allow cargo transport. The Kankoh-Maru is named S-1 in USAF service.

The authors are well aware of the unlikelihood of a Japanese passenger transport rocket being bought and operated by the USAF! Thus it is important to remember that this is an illustration only. However, the "S-1" is currently the only reusable VTOVL rocket of which the design has reached a sufficiently advanced stage to enable the preparation of a Flight Manual. Perhaps in future the USAF may need regular transport to and from an orbiting station. For illustrative purposes, there are determined to be 3 versions of the S-1 (see Appendix):

An experimental version, equipped with ejection seats for the crew, and with no provision for carrying passengers. This version is included to illustrate those operations relevant to flight test, functional check flights, and especially, those relevant to a smaller X- SSTO, which is herein assumed to have preceded the S-1's construction.

The basic model, assumed to have been in service with the USAF and airlines for a small number of years.

A newer, more advanced model, incorporating a few improvements such as avionics upgrades, replacement of the Al-Li LOX tank with a lined composite tank, and replacement of all hydraulic systems in the vehicle with electro-mechanical actuators.

Based on the accompanying Flight Manual (Appendix), in this section typical procedures are considered for the different phases of a routine passenger flight to and from orbit to rendezvous and dock with an already orbiting craft, analogous to a present-day airline flight. It is hoped that this will provide a realistic image of Kankoh-maru's airline-like operation, and show that, in contrast to existing launch vehicles, it is realistic to envisage Kankoh-maru being operated by airlines in a manner essentially similar to the other aircraft in their fleets.

In modern commercial airliners automation has advanced so far that for most phases of a flight active control inputs by the pilot are unnecessary. Despite this, the pilot plays a number of technically and legally critical roles, including deciding the flight plan, deciding to commit to take-off, and carrying the various legal responsibilities of captain of the vehicle for the duration of the flight.

It seems likely that the case of flights to or from orbit will be broadly similar to an airline flight, except that the legal regimes involved will be different from a terrestrial flight. It has already been recognised that a major requirement for the establishment of passenger space travel services is the appropriate revision of space law, which currently comprises mainly inter-governmental treaties negotiated during the cold war (3). The definition of the legal powers and duties of the pilot of a commercial passenger-carrying spacecraft will be an important part of the new structure of law that is required. It is anticipated that the new legal framework will be based primarily on existing international commercial law, including both aviation law and maritime law, but such a discussion is beyond the scope of this paper (3).

As explained in the Flight Manual (Appendix), the crew are assumed to comprise a pilot who is the captain, and a flight engineer. The role of flight engineer has been removed from the most recent generation of airliners, but was routine until recently. Airline captains have a Private Pilot's license (PP), and Commercial Pilot's license (CP) in addition to their Air Transport Pilot's license (ATP) and "type-rating" for the airplane that they are currently piloting. Thus it is assumed that Kankoh-maru pilots will have these three licenses in addition to a Space Transport Pilot's license (STP) and appropriate "type rating".

Like the crew of commercial aircraft, on signing in before the flight, the Kankoh-maru crew will certify that they are alcohol- and medication-free, and that they satisfy a number of other safety conditions such as restrictions limiting the number of hours that they may fly during a given period of time. Although flights from Earth to low orbit can take as little as a few minutes, and flights from low orbit back to Earth can take less than an hour, there will be complex constraints on the timing of flights imposed by the relative positions of the take-off site, the orbital destination (such as a hotel), and the landing site, which will tend to make flights last several hours or more. Orbital mechanics will thus add new factors to the planning of crews' work-schedules, as well as to the scheduling of flights to and from orbit.

As in modern aircraft VTOVL launch vehicles will have advanced health-monitoring systems. However it is assumed that, following aviation practice, the flight crew will be required to check the vehicle condition, both exterior and interior, as a further safety precaution. Due to the different characteristics of cryogenic propellants from aviation fuel, the external walk-around may be made prior to propellant loading, with video cameras being used for further continuing checks after propellant loading.

One of the crew's major duties before take-off will be to check the flight plan. As for airliners this will involve checking a) the route, b) the weather, and c) the propellant load.

Route. The orbital path of the target craft, such as a hotel, is fixed, and so its position relative to both the take-off site and the planned landing site are determined by its orbital parameters. Consequently the route and timing of both the outward flight and the planned return flight are irrevocable from take-off, even if the return flight is scheduled for some hours later (except of course in the case of an unplanned change of schedule). Consequently the details, including particularly the planned time and place of re-entry and the subsequent atmospheric flight and landing will be confirmed before take-off. Contingency plans for re-routing will also be made prior to take-off in case of subsequent cancellation or postponement of the return flight to a particular destination airport. These are details which the dispatchers of Kankoh-maru flights will consider in planning them so as to optimize their company's orbital flight schedules.

Weather. Since Kankoh-maru flies in the atmosphere for only a few minutes, relevant weather conditions will be more predictable than for airline flights lasting several hours. However there will be some additional complexities. First, relevant weather conditions for flights to or from orbit will include a wider range of information, specifically including the condition of the upper atmosphere (the height of which varies with the level of solar activity) and any solar storm activity which could expose Kankoh-maru passengers to higher than normal radiation levels. Second, since the return flight will also be scheduled before take-off, even though it may be several hours later, and since weather cannot be predicted with certainty over periods of hours, there will always be a possibility of having to cancel, postpone or re-route a return flight. Third, the return flight may be scheduled in a distant part of the world, as little as two hours in the future, and so detailed weather forecasts will be required on a global basis, at least for the main re-entry corridors to the international airports which will operate Kankoh-maru flights.

Propellant Load. Kankoh-maru's propellant load will depend on the number of passengers, the total payload mass, and the route, both outward to orbit and return to Earth. As discussed above, it will not be economical for Kankoh-maru to carry excess propellants, in contrast to scheduled airline flights which carrry a reserve of fuel sufficient to reach an alternate landing site in the event of a problem preventing them landing at the intended site. However, Kankoh-maru will not use an aircraft runway, and in emergency could land in almost any open area. Consequently the major reason for airliners' carrying fuel reserves, namely in order to guarantee the ability to reach another airport with a runway several kilometers long, does not apply to Kankoh-maru.

As passenger services to and from orbit grow in scale and maturity, it will probably become possible to purchase propellant supplies in orbit. In particular, from an early stage it will probably be possible for Kankoh-maru to obtain supplies of liquid oxygen from orbital facilities which it visits, since hotels will hold large reserves of liquid oxygen for its air supply. Since liquid oxygen represents some 85% of the mass of Kankoh-maru's propellants, it may become normal for Kankoh-maru to purchase liquid oxygen in orbit to use for its de-orbit and landing maneuvers since this will add greatly to the flexibility of its operations. (The economics of such orbital re-fueling operations will depend on the cost-reduction achievable by delivering liquid oxygen to orbit using tanker vehicles, and on the cost of storage in orbit. These will determine whether it can be supplied at a price lower than the cost of carrying it to orbit on each passenger flight.)

Before final preparations for take-off are made the captain will contact Air Traffic Control to request clearance for take-off at the scheduled time. As discussed above, this will be somewhat different from an aircraft clearance, since Kankoh-maru will use a different, almost vertical departure route, and clearance will not be subsequently over-ridden by other aircraft delays or requirements, except in case of emergencies.

As described in the Flight Manual (Appendix) the main engines will be started in sequence in idle mode at some 5% of full thrust, while Kankoh-maru is connected by quick-disconnect fuel lines to the Departure and Landing Facility (DLF) as described in (2). The scheduled take-off time will be the time at which the orbital plane of the hotel with which Kankoh-maru is to rendezvous passes through the take-off site. (The hotel's orbit plane is effectively fixed relative to the distant stars, and the Earth rotates relative to it at slightly more than 15 deg/hour.) It is very inefficient to launch into a different orbit plane, and so launching into the same plane enables Kankoh-maru to reach its orbital destination using the minimal quantity of propellant (4). The engine starting sequence will begin at a time before the scheduled take-off time that allows the nominal length of time for engine start. Once the captain confirms that all engines are running normally the decision to take off will be made, and the engines will be accelerated together to 100% thrust.

The decision by the captain of an airliner to take off is safety-critical because of the very short period of time available to check that all the aircraft's engines and systems are operating normally after the aircraft has reached flying speed, but before there is insufficient distance for the aircraft to brake to a halt safely, due to the physical limitation of the length of runways.

By contrast, the captain of a Kankoh-maru flight to an orbital destination such as a hotel will not have such a safety-limited time in which to confirm normal engine operation before the take-off must be aborted. At full thrust the main engines will produce some 1.4 times the weight of the vehicle. Consequently at any moment before the engines reach some 70% of full thrust, they can be shut down and the vehicle will remain safely at the DLF. Thus the final decision to take off will be taken as Kankoh-maru's main engines accelerate together through the critical value of 70% thrust, for which Kankoh-maru's on-board computers will monitor the performance of all 12 engines in real time. Once the engines exceed 70% of full thrust Kankoh-maru will leave the DLF. Subsequent to that time a decision to abort the flight, either due to an engine failure or for some other reason, will involve following the procedures for the appropriate flight abort mode described in the Flight Manual (Appendix).

There are two main constraints on the time-window during which Kankoh-maru can take off. Depending on the relative positions of the take-off site, the hotel orbit plane and the hotel's position within the orbit, and on the time available for approach and docking (which will depend on the subsequent flight schedule), Kankoh-maru will have a relatively narrow take-off "window" during which it can reach its orbital destination. Delay beyond the limit of the launch window would prevent Kankoh-maru from being able to reach its scheduled orbital destination, leading to cancellation of the flight. It should be noted that this condition is not independent of the propellant load carried, since a larger quantity of propellant carried to orbit would permit Kankoh-maru to make a somewhat larger orbital plane change, thereby extending its launch window, though not by very much (4).

A second constraint is the cost and speed of propellant use. Airliners use fuel costing about $100/ton at a rate of about 10 kg/second, or approximately $1/second. By contrast, Kankoh-maru uses propellants costing approximately $1,000/ton at a rate of about 2 tons/second when all engines are at full thrust. Thus running all engines at idle thrust of some 5% of full thrust will cost some $100/second, while running them at full thrust will cost some $2,000/second. Thus delaying take-off for one minute with engines running even at idle would cost some $6000. There is therefore a much stronger incentive for operators to minimize unnecessary engine running-time in Kankoh-maru's case than for an airliner.

For short-distance ferry flights, Kankoh-maru's take-off propellant load and take-off thrust will be only a fraction of the maximum, and so there may not even be a need for a dedicated flame-trench. In such cases Kankoh-maru could take off from sites other than specialized take-off facilities.

Once Kankoh-maru has taken off, its engines will, like a modern airliner, follow a pre-programmed sequence, the main difference being that powered flight to initial low orbit typically takes only 5-6 minutes, in contrast to the several hours of most airline flights. As the propellants are used up Kankoh-maru will get lighter and, according to the current vehicle configuration, after some 100 seconds the 4 booster engines will start to be throttled back progressively to prevent the acceleration increasing beyond the nominal maximum of 3 G. After about another 20 seconds the 4 booster engines will be shut down. (Note that they will be shut down in a manner appropriate for later restarting for de-orbit and landing.)

In order to increase the propulsive efficiency of the 8 sustainer engines at high altitude, their nozzle-extensions will be extended. They will also be throttled back progressively to keep the vehicle's acceleration below 3 G. As they approach their minimum thrust levels, four will be shut down. Finally, once Kankoh-maru reaches its planned initial orbital altitude the last two pairs of main engines will be shut down. As for a modern jet-liner, all these steps will be sequenced automatically by the flight control computer following the flight plan. The flight crew's main task will be to monitor the on-board computers and to be ready for abnormal conditions and emergencies.

Flights of the S-1 will be less routine, and so although the flight control computer of the S-1 will have similar capabilities, it may not be used on all flights. While commercial pilots will be required to practice regularly on simulators, military pilots will be required to have regular practice of actual direct control of the vehicle, as is normal in military aviation but is not practical for airline service. It is also possible that some versions of the S-1 may be designed to operate at up to more than 3 G acceleration. This would have the advantage of economizing on propellant use and enabling a larger reserve of propellant to be carried to orbit where it could be used for maneuvering.

Once Kankoh-maru is in its planned initial orbit, the captain will establish a slow rate of roll in order to give all passengers the full range of views available (1). For a typical flight involving docking with an orbiting hotel for transfer of passengers, the pilot will need first to follow an appropriate rendezvous procedure. The rendezvous procedure used will vary between flights, depending on the latitude of the take-off site, the inclination of the orbit, and the phase difference between the target hotel and Kankoh-maru at take-off. Typically the procedure will involve flying in an orbit at 200 km altitude for some hours in order to "catch up" with the hotel, followed by a sequence of several engine burns to raise Kankoh-maru's apogee and perigee, and possibly also to adjust its orbit plane (4).

As Kankoh-maru approaches the hotel it will enter the hotel's Traffic Zone, and the captain will follow the local traffic control rules under instruction of the hotel traffic controller. These rules have yet to be decided; this will be the responsibility of the hotel operators, but the system that has been conceptualized to date comprises a series of nested zones in the same orbital plane as the hotel (5). According to this proposal, the "Transportation Zone" covers the entire toroidal region extending some tens of kilometers above, below and to the sides of the hotel orbit, and around the orbit; the "Forward and Backward Co-orbiting Zones" will be regions within this toroid up to a few thousand kilometers ahead of and behind the hotel and within perhaps 10 kilometers vertically or laterally; and the "Proximity Zone" will be a spherical region centered on the hotel with a diameter of some 10 kilometers (5). For traffic control purposes there will be standard approach and departure routes within these zones. The legal issues involved in defining such traffic rules are somewhat complex, and there will be operational advantages in having different commercial facilities sharing the same orbit, as discussed in (6).

In general, performing final approach from vertically below the hotel along the orbital radius vector, rather than horizontally along the velocity vector, has a number of advantages, including being fail-safe, and avoiding thruster plume inpingement ("pluming") on the target facility (4). When the flight schedule and propellant supplies allow it, Kankoh-maru may perform a "fly around" of the hotel in order to give passengers scenic views of the hotel against the backgrounds of Earth and space (4). The time available for such manoeuvres will depend on the flight-schedule, and particularly on the time until Kankoh-maru's subsequent return flight from the hotel.

Flights of the S-1, rather than being optimized for minimum cost and minimum propellant use, like scheduled commercial flights, may more frequently involve significant orbital manoeuvring which will require additional propellant. Carrying a minimal payload an S-1 could carry an extra 5-6 tons of propellant. In addition, provided that it was able to refuel in orbit by docking with a propellant supply facility, the S-1 would be able to use the several tons of propellant normally needed for de-orbit and landing for orbital maneuvering. 12 tons of propellant would give an S-1 with a 50-ton empty mass a maneuvering capability ("delta V")of approximately 900 m/sec.

After a routine approach and docking, Kankoh-maru's flight crew's tasks will depend on the services available at the facility, which will vary between facilities and also over time. At the earliest stage of passenger accommodation services in orbit, capabilities such as maintenance at these facilities will be minimal. However, like the services at a company's airport terminal, they can be expected to grow and develop as the scale of orbital travel activities grows, due to the operational and business advantages of doing so.

Thus, at a minimal facility, the Kankoh-maru's crew will treat the visit as a stop-over on a continuing flight, rather than as the end of a one-way flight. In this case they will not formally shut the vehicle down, and in the case of a short stay they may not even leave the cockpit, the captain retaining legal responsibility throughout the stay. As for airliners, any anomalies will be recorded in the vehicle's log book, and since real-time global communications will surely be available, the vehicle condition will be able to be confirmed with company engineering staff on Earth or elsewhere in orbit. In the event of a problem requiring capabilities or parts not available at the orbital destination, these could be delivered on another flight, either from Earth or from another orbital facility.

Even at such an early stage the legal relations between the captain of the Kankoh-maru and an orbital hotel manager will have considerable significance as the passengers pass from the captain's "jurisdiction" to that of the hotel manager for the duration of their stay. The details of such legal arrangements have yet to be decided, but it will be necessary for them to be finalized before investors will be prepared to undertake the financial risks involved in building orbital facilities (7). Such matters will also need to be agreed internationally.

As orbital hotels become more advanced, the range of services which they will offer to Kankoh-maru will grow to include such activities as cabin cleaning, cargo loading (as appropriate), and exchange of members of the flight crew. On-orbit maintenance capabilities will also expand, particularly checks (perhaps using remote diagnosis by the ground crew), change-out of LRUs, and others. As passenger travel services grow and mature, the growth of orbital maintenance capabilities is likely to involve specialization at different facilities, creating the possibility of economically obtaining repairs at another facility (particularly if it is in the same orbit). As discussed above, at some stage also propellant supplies, particularly liquid oxygen supplies will become available in orbit.

In the early stages of orbital travel services, the return flight may be treated as the second leg of a single flight. Nevertheless, before departure on the return leg the flight crew will re-check the flight plan, involving the route including confirmation of the scheduled airport landing-slot, weather conditions en route from orbit to the landing site, and the propellant load, which will depend on the number of passengers, the overall payload, and other conditions.

As passenger space travel services mature and propellant supplies, or at least LOX supplies, become available in orbit, planning of propellant use will become easier. It will even become possible for Kankoh-maru to sell LOX which may be surplus to its requirements for the return flight, to the hotel. Until that time, Kankoh-maru's return payload will be limited by the propellant remaining in its tanks, and by the vehicle's return flight performance specifications.

The need for, and appropriate way to perform an external check of Kankoh-maru when docked to an orbiting hotel will depend on the design of the hotel in question, but much of the vehicle should be visible from inside the hotel. However, performing external operations requiring space-suits will be possible at any orbital facility, and so, if considered necessary, it will be possible to make a closer external inspection of Kankoh-maru. At an earlier stage of development, when the period of docking is short, it is less likely to be considered necessary to make a complete external inspection. Later, particularly after a longer period in orbit, it will presumably become routine to make a complete external check.

Before departure the pilot will also re-contact Air Traffic Control at its destination airport for confirmation of landing clearance. Permission to depart will also be required from the hotel traffic controller who will be responsible for routing traffic within the vicinity of the hotel. Although traffic will be much less than at airports, orbital hotels are likely to have daily arriving and departing flights of both passenger and cargo vehicles, growing to several flights per day. In addition, many hotels may choose to operate in the same orbit, making traffic control an important function.

It should not be necessary to perform a test-firing of the main engines before departure from the hotel, because of the engine redundancy on the return flight: only two engines are needed to operate at half-thrust in order to decelerate and land safely. Test-firing the main engines while attached to the hotel would in any case be impractical, although it could possibly be used for orbit-raising of the hotel if this had been designed for. For some facilities this might be a cost-effective way of maintaining orbital altitude, without requiring an autonomous propulsion system with propellant supplies and other equipment. (It is also likely that at some stage tethers will come to be used for departure, thereby economizing on propellant use both for de-orbit and for reboosting hotels in their orbit.)

Departure from the hotel, other than for the most basic facilities, will be assisted by giving Kankoh-maru a substantial initial impulse using either a spring or compressed air. This has the advantage of being deliverable using electric power which can be generated using sunlight in orbit, rather than using propellants brought to orbit from Earth. It also reduces the quantity of propellant that Kankoh-maru would need to carry to orbit, thereby improving its payload capacity, and would also reduce the possibility of causing damage to the outside of the hotel through "pluming".

After departure, Kankoh-maru will head towards the site of its planned re-entry burn, following rules for spacecraft movements within the traffic zone around the hotel. When its time-schedule permits, Kankoh-maru may follow a trajectory including a fly-around of the hotel in order to give passengers interesting views. This will need to be timed precisely in order for Kankoh-maru to arrive accurately at the site of its planned engine-burn for de-orbit.

Kankoh-maru is designed to have a nominal cross-range capability of 200 kilometers, since with modern navigation systems this is considered sufficient for scheduled commercial passenger operations. The orbital position and timing of the firing of two of the vehicle's booster engines in order to de-orbit will tightly determine its point of atmospheric re-entry, and thence its atmospheric flight envelope and possible landing sites. Modern navigation systems make it simple to achieve the precise positioning and timing required for the engine burn to minimize propellant use. However, it would be possible to reach the same landing-site from a less optimal re-entry position by using more propellant.

When on-orbit propellant supply facilities become available they will give Kankoh-maru important additional capabilities. First, by replenishing its propellants in orbit, Kankoh-maru would be able to perform in-orbit maneuvering, including a significant orbital plane-change if necessary, thereby enabling it to visit other orbital facilities. By carrying a propellant load greater than the minimum needed through re-entry, Kankoh-maru could also re-enter with sufficient propellant supplies on board to achieve extended cross-range before touch-down, although this will be limited by the extra re-entry heating that this would cause. Further, because Kankoh-maru has no wings, it is a very efficient design for operations in space: for example, refilling Kankoh-maru's propellant tanks in orbit would give it the delta-V capability to make return flights to the lunar surface, for which winged vehicles are not suitable (8).

After re-entry, two booster engines will be used for deceleration and landing. As discussed above, Kankoh-maru will need special air traffic arrangements due to its inability to hover for a significant length of time before landing. However, since Kankoh-maru does not need a runway, there are no realistic emergency conditions in which it would be required to travel a significant distance to an alternative landing site, as aircraft are required to be able to do. Furthermore, given that the 55-ton nominal landing weight is only 10% of the take-off weight, the engine thrust required for touch-down is so low that landing on ordinary concrete or even unprepared sites could be acceptable.

Using existing GPS navigation capabilities and short-term weather forecasts it is expected that Kankoh-maru's flight control system will enable it to land within a range of 50 meters of its target landing-point. It can be anticipated that 10 years from now, when Kankoh-maru could be entering commercial service, the capabilities of these systems will be even greater than they are today.

After landing, the flight crew will shut the engines down and Kankoh-maru will be towed to the DLF for connection to the propellant lines and preparation for its next flight. The flight crew will complete a shutdown checklist, and note any maintenance items in the log book. On leaving the vehicle they will pass responsibility for Kankoh-maru over to the ground crew, and it will receive maintenance as required following a standard preventive maintenance program developed during the initial test-flight program and subsequently improved during flight operations.

Aircraft maintenance is based on a hierarchy of checks - a "trip check" after every flight, a "layover check" every day, a "service check" every N hours of operation (depending on aircraft type), and a "periodic check" varying in severity depending on the aircraft's accumulated flight hours. In addition there are periodic major overhauls which effectively return the aircraft to "as new" condition. A similarly sophisticated system of preventive maintenance will be developed for Kankoh-maru based on operating experience. According to current plans, 4 Kankoh-maru test vehicles will be manufactured and test-flown 1,200 times over a period of 2 to 3 years (9). During both the earlier engine development and test program and the flight test program, appropriate maintenance procedures and TBOs for the various components and subsystems will be developed progressively. Production line manufacture will then produce 8 vehicles per year which will be operated in fleets by airlines. This will quickly generate a large database of cumulative experience. With today's advanced manufacturing and computer systems, international operations and real-time communications, the learning process will be faster than it has been for the aircraft industry, and it should take only years rather than decades to reach very high reliability. (NB by following aviation procedures, a very high level of safety can be achieved even in the earlier phases of the service when reliability of individual parts of the system is less than in airliners today.)

The key to the realization of routine, low-cost transport services between Earth and space using the Kankoh-maru reference vehicle will be to achieve low-maintenance operations, based on design for long lifetime and standardized preventive maintenance procedures as used for aircraft, particularly for the rocket engines, cryogenic propellant tanks, and thermal protection system. In order to achieve this it is essential that designers of reusable launch vehicles adopt the fundamental operating philosophy of aviation, even though it is unfamiliar to them. Only in this way can they start to tap the passenger transportation market, which alone offers the prospect of generating large-scale commercial operations - as it has for aviation.

Details of the operation of Kankoh-maru will depend on details of the final vehicle design, particularly of the engines and the flight control system. VTOVL rockets differ from aircraft in a number of ways, and such matters as the most appropriate interface between Kankoh-maru's pilot and the control-system remain to be determined. However, many such aspects are likely to draw heavily on aircraft experience.

Operating experience of piloted, reusable rocket vehicles began to be accumulated during the second world war, continued with the US "X"-planes in the 1940s-60s, and then with the RATO aircraft operated in several countries during the 1950s. More recently the unpiloted McDonnell Douglas DC-X and DC-XA vehicles started to accumulate experience of VTOVL rocket operations. There is an urgent need to continue accumulating operating experience of piloted reusable rockets on a larger scale, in order to achieve the same level of reliability and safety in space transportation as is routine in aviation.

AC	Alternating Current
ADI	Attitude Direction Indicator
AGL	Above Ground Level
ATC	Air Traffic Control
ATPL	Air Transport Pilot's License
BIT	Built-In Test
BPU	Bus Power Unit
CC	Ground Crew Chief
CP	Copilot (ground)
CP	Commercial Pilot
DC	Direct Current
DLF	Departure and Landing Facility
EFI	Electronic Flight Instrumentation
ELV	Expendable Launch Vehicle
FADEC	Full-Authority Digital Engine Controller
FE	Flight Engineer
FOCC	Flight Operations Control Center
GHe	Gaseous Helium
GLOW	Gross Lift-Off Weight
GPS	Global Positioning System
GPWS	Ground Proximity Warning System
IFF	Identity, Friend or Foe
INS	Inertial Navigation System
GN	Gaseous Nitrogen
KIAS	Knots Indicated Air Speed
LE-9B-3	augmented expander cycle, LH2/ LOX booster engines
LE-9S-3	augmented expander cycle, LH2/ LOX sustainer engines
LH2	Liquid Hydrogen
LOX	Liquid Oxygen
LRU	Line Replaceable Unit
N	Navigator (ground)
PP	Private Pilot
RATO	Rocket Assisted Take-Off
RCS	Reaction Control System
RPM	Revolutions Per Minute
SID	Standard Instrument Departure
SIF	Selective Identification Feature
SSTO	Single Stage To Orbit
STAR	Standard Terminal Arrival Route
STP	Space Transport Pilot
TA	Turbo-Alternator
TBO	Time Between Overhauls
TIT	Turbine Inlet Temperature
TP	Transport Pilot
VTOL	Vertical Take-Off, Vertical Landing
VVI	Vertical Velocity Indicator

1. K Isozaki et al, 1994, " Considerations on vehicle design criteria for space tourism", IAF paper no IAF-94-V.3.535
2. M Nagatomo et al, 1995, "Study on airport services for space tourism", AAS Vol 91, pp 563-582
3. P Collins, 1996, "The regulatory reform agenda for the era of passenger space transportation", Proceedings of 20th ISTS, Paper No 96-f-13
4. T Williams and P Collins, 1997, "Orbital considerations in Kankoh-maru rendezvous operations", Proceedings of 7th ISCOPS, AAS in press
5. M Nagatomo et al, 1984, " Orbital operation of co-orbiting spacecraft with space station", IAF paper no IAF-84-42
6. P Collins and T Williams, 1986, "Towards traffic systems for near-Earth space", 29th Colloquium on the Law of Outer Space, IISL, pp 161-70
7. C Lauer, 1996, " Analysis of alternative governance models for space business parks", Engineering Construction & Operations in space, ASCE, Vol 1, pp 177-185
8. G Hudson and M Hyson, 1988, " A single-stage vertical takeoff and landing space transport for lunar settlement and establishment and resupply", Proceedings of 2nd Lunar Bases and Space Activities of the 21st Century Conference, Houston
9. K Isozaki (ed), 1997, " Kankoh-maru cost estimate and development plan", Japanese Rocket Society (in Japanese).