Man versus robot

“Man is the best computer we can put aboard a spacecraft… and the only one

that can be mass produced with unskilled labor.” – Wernher von Braun

Will a future hypersonic plane have pilots on board or be fully automatic? Nowadays aerodynamics and rocket propulsion are fairly well understood and can be accurately modeled and simulated. As a result, the areas where direct pilot intervention may be needed due to unexpected behavior of an aircraft/spaceplane are rapidly decreasing. This is especially true for vehicles that have a fixed, pre­determined trajectory such as missiles and launch vehicles. So-called Unmanned Aerial Vehicles have become important operational military assets, and these aircraft are steered from the ground or fly their missions completely autonomously as aerial robots. It is therefore likely that future space planes will be flown by a computer under human supervision from the ground rather than directly by a human pilot, particularly as hypersonic vehicles tend to be aerodynamically unstable and therefore require sophisticated avionics for efficient and safe control. For instance, Skylon is to fly automatically; any astronauts to be transported into orbit will be housed inside its payload bay.

Especially on a satellite launch vehicle with relatively little margin for errors and malfunctions, operating without pilots results in a simpler and thus cheaper design; a crew requires a comfortable cabin with regulated pressure and temperature, requires to have an escape capability if the spaceplane is less reliable than a regular aircraft (which rocket vehicles invariably are) and requires higher safety margins to be built into the design. Not having any people on board potentially makes the vehicle less expensive and saves weight and space that can be used for more payload. Moreover, catastrophic failures have less grave consequences; compare the dramatic aftermaths of the losses of the Space Shuttles Challenger and Columbia with those of the many but almost forgotten failures of unmanned expendable launchers.

However, on many occasions having an exceptionally skilled pilot on board saved the X-15 and earlier rocket aircraft. So any fully automatic flight control system on a versatile hypersonic aircraft intended for various types of missions must be smart and capable of reacting very rapidly to unexpected situations and emergencies. That may be difficult, as programming a computer for unforeseen events is near to impossible whilst the human brain excels at improvisation (although developments in so-called neural networks may result in self-learning computers that can quickly react to new situations). And what if a spaceplane carries astronauts onboard? Even if they are not flying the vehicle themselves, the aircraft will still need to incorporate the additional equipment and various reliability enhancing redundancies that an unmanned vehicle can do without. Would it be acceptable for them to ride into space in a fully robotic hypersonic launcher? Or would a pilot with a manual override capability be required, as for astronauts launched on current expendable rockets like the Soyuz and even the Space Shuttle, if only for psychological reasons? The impact on the design would be limited if one of the transported crew could fly the vehicle in an emergency, in order that no additional seat need be assigned to a pilot.

Talking of people on board spaceplanes and rocket planes in general, what about vehicle safety? The early rocket propelled aircraft like the Me 163 were extremely hazardous. Four pilots died and two were severely injured during the X-l, X-2 and X-15 programs and there were also many less serious accidents. Of the 16 individual airframes involved, 10 were completely or largely destroyed in accidents: not a very good safety record given that the X-planes only made a total of some 415 flights, a total that can be readily accumulated by a single airliner in 6 months of operations. Does this mean that rocket planes are inherently dangerous and hence ought never to be used for suborbital space tourism and/or mass transportation into orbit? Surely we have learned much about high-speed, high-altitude flight since those days, and rocket propulsion has also greatly matured. Suborbital flight in particular, benefits not only from the experience gained from the experimental rocket planes but also from high-performance jet aircraft in general.

Furthermore, whilst the high losses among pilots flying rocket planes may appear high today, they were not particularly exceptional compared to the accident rate in experimental aviation and the general testing of prototype aircraft. In the late 1940s and the 1950s test pilot loss rates in the US were in the order of one per week. And crashes of military jets in operational service occurred frequently. Nowadays crashes and aircraft explosions are very rare, even for new types, so there is no real reason to expect suborbital rocket aircraft like SpaceShipTwo to suffer from anything like the loss rates of early jets. However, a suborbital launch is certainly more hazardous than a regular airline flight, and orbital spaceflight even more so. In part this is due to the extreme speeds, altitudes and temperatures involved, in combination with the need to keep the vehicle as light as possible, and in part due to the still experimental nature of human spaceflight. At the time of writing, the number of crewed space missions is less than 290, well below the number of planes in the air on a typical day. There have been even fewer suborbital rocket plane flights into FAI-certified space. Indeed, only two X – 15 flights and the recent three missions of SpaceShipOne ascended above the milestone altitude of 100 km (62 miles), and another eleven X-15 flights exceeded 80 km (50 miles). In today’s world of health and safety regulations, the relatively low trustworthiness of rocket vehicles is certainly a business risk. People have come to expect that even radically new aircraft will not kill anyone, and that suborbital space tourists riding rocket planes should not feel that they are putting their lives on the line. On the other hand, perhaps it is the risk that provides the sense of adventure.