The fourth Duke of Edinburgh's Lecture was presented by Professor P. M. S. Blackett, C.H., P.R.S., at the nineteenth Annual General Meeting, held in the Royal Geographical Society's house on 27 October 1965, with the President in the Chair. The President introduced the lecturer as one of the most eminent physicists of our time. Over the last thirty years he had held a number of very distinguished appointments, and was until recently Professor of Physics at the Imperial College of Science and Technology. He is now deputy chairman of the Ministry of Technology's Advisory Council. In 1948 Professor Blackett was awarded the Nobel Prize for Physics. A summary of the main points covered is given below.

Of all man-made objects, satellites are unique in their ability to ‘see’ large regions of the Earth at one time. Because of this ability, they can serve as relays to provide the advantages of line-of-sight communications between widely separated points on the Earth. They have another useful characteristic also—their locations in space can be determined to high accuracy, hence they can be used as position references for navigation or surveying.

1. Types of presentation. The choice of the method of presentation used by a radar operator is probably the most important of the various alternatives open to him for the most effective keeping of his watch, because they are all different, each has its advantages and limitations and one of them may be best suited to the circumstances and to the mentality of the observer himself. In what follows, the anti-collision use is principally in mind.

In various papers published recently in the Journal and elsewhere, examples have been given in which the theoretical normal or gaussian distribution of errors does not correspond closely with the distributions found in practice; in particular the frequency of large errors is commonly underestimated by the normal law of errors. This discrepancy is discussed by Anderson who suggests that an exponential distribution may give a more realistic estimate of the frequency of large errors. A closely associated problem is that of determining the frequency distribution of the largest values occurring in a number of samples taken from some parent population. This problem has been solved theoretically for an important class of distributions and it is the object of this article to draw attention to this work and to show how it can be simply applied to practical problems, albeit in a rough and ready manner, without going far into the statistical details with which the subject abounds.

1. Introduction. Nautical charts of many areas of the world carry notations adjacent to natural marine hazards such as ‘position doubtful’, ‘existence doubtful’, ‘position approximate’ and ‘reported (date)’. In spite of the development of precision survey and navigation systems, many of these notations must continue to be carried until the positions of the hazards are accurately determined or their existence disproved.

The expanse of the oceans obviously precludes costly systematic hydrographic surveys in all areas, particularly in remote offshore areas. Although international cooperation in the exchange of hydrographic information assists in meeting modern charting requirements, the task of charting the seas may never be complete.

The following papers were presented at a meeting of the Institute held in London on 12 March 1965. Mr. W. P. Davies (Polhemus Associates) first outlines the height measurement requirements for a supersonic transport aircraft. The other speakers review the available techniques: Mr. R. Whalley (Ferranti) dealing with inertial techniques, Mr. B. Cape (S. Smith & Sons) with pressure measurement techniques and Mr. W. L. Garfield (Standard Telephones and Cables) with radio altimetry. Mr. G. E. Beck (The Marconi Company) describes a recently developed radio altimeter.

1. Introduction. In outlining requirements for height measurement in supersonic transport aircraft it should first be said that not all the requirements are special to supersonics. Some of the requirements are the same as already exist, or will exist in the 1970's, for subsonic aircraft. The main problem in supersonic aircraft will be to maintain the same accuracy over increased measurement ranges of height and height rate and in an environment of higher speeds and temperatures. Where-ever possible then, the requirements are of the same standard expected for subsonic aircraft of the 1970's. More stringent requirements are stated only where they are considered necessary; there are enough real problems in the advance to civil supersonic operations without creating artificial ones.

Inertial navigation techniques have been applied as a means of improving the height information in military aircraft for some time. It is almost certain that the supersonic transport will use inertial techniques for plan navigation and it is interesting to see how far it is worth extending this to include height, and also height rate, information.

It is convenient to start by looking at the fundamental principles of inertial navigation as applied to height measurement; Fig. 1 shows the essential features. There will be an accelerometer measuring vertical acceleration; this must be maintained approximately vertical and can be very conveniently mounted on the main inertial platform. The output of the accelerometer contains a number of unwanted terms for which corrections have to be made. The expressions are well known and are explained in the Appendix.

From an examination of the outline requirements given by Mr. Davies, it is apparent that pressure measuring techniques can be used in all stages of a flight although it must be acknowledged that some of the tolerances he has called for cannot be met. The advantage of the pressure technique is that it uses the Earth's atmosphere which is experienced by all airborne vehicles and so provides a common datum for height or altitude measurement. It should be noted that wrong impressions can be got from the present pattern of accepting altitude or height in feet as a fixed physical measurement. A pressure sensor measures the pressure and, as the millibar is a widely accepted unit of pressure in aeronautics, it would really be more sensible to express aircraft altitude and separation values in millibars.

In the terminal area and for the approach and landing phase, radio provides the most appropriate method of height measurement. That is to say that the altitude reference frame is switched from barometric to terrain clearance measurement. It is at first a little surprising to see from Mr. Davies' requirements for this phase that the accuracies are in line with those currently in use in subsonic jets. But consideration shows that this is because the approach speed and general handling characteristics of the S.S.T. must approximate to those of subsonic jets in the approach and landing phase. Mr. Davies has also stated that automatic landing is essential for the S.S.T., so that the picture of requirements for radio altimetry in the approach and landing phase is fairly clear.

This contribution is somewhat aside from the flow of the discussion so far. Mr. Garfield has dealt with the general features of radio altimetry and I am going to describe only one version of a radio altimeter in a little more detail. What I have to say is also a progress report in continuation of an earlier note published in the Journal; at that time the possibility of obtaining altitude as additional information from an airborne doppler navigator was discussed, and it was concluded that an accuracy of ± 5 per cent ± 2 ft. over the range of 10–500 ft. was possible. It was emphasized that the work was in an investigational stage, and no specific design was described. Today, over five years later, I can report on the flight trials of an engineered equipment using these principles. The long gap is not a true measure of the total time from design study to a unit in production, but the hazards of either commercial or military supported development programmes can extend the time in a manner which is all too familiar to many people.

1. Hovercraft Environment. With the exception of the sidewall craft which is unlikely to achieve speeds in excess of 30 knots, all hovercraft depend on their ability to rise clear and free from water drag in order to achieve their high economical speeds. Nevertheless a hovercraft will tend to conform with the surface over which it is travelling. A rough sea produces a rough ride which at speed produces a motion not unlike that of a car going over a series of hump bridges with vertical accelerations in the order of 1½ g—one needs to be strapped into a seat. The most efficient means of propulsion is achieved by using air screws; these are noisy and necessitate crew members using an intercom. Control surfaces, rudder and airscrews are aerodynamic and to reduce air drag the hovercraft is likely to be low in silhouette and give poor all-round vision to the command. The biggest single factor is drift; being aerodynamic and free of water drag, the hovercraft is completely wind borne and in moderate beam winds can track with as much as 45° of drift angle on and turns out of wind can be alarmingly large.

This and the following paper by Mr. Attwooll are the first of a series which will describe the work of the Royal Aircraft Establishment on various aspects of air collision risk and traffic control. Mr. Reich's paper will be presented in three parts, the last two of which deal with the analytical techniques and how they are applied. Mr. Attwooll deals with methods of costing the effects of air traffic control deviations and shows how these costs are related to the size of separation standards. He discusses systems for both subsonic and supersonic transports.

1. Objectives. The main task of air traffic controllers is to plan traffic flows so that aircraft are allotted sufficient separation to absorb not only systematic differences in speed but also the imperfections of navigation and piloting, which we term flying errors. To this end, they usually work with three separation standards, to be applied in, respectively, the along-track, across-track and vertical dimensions of space. The separation standard for any one dimension is the minimum intended (i.e. planned) separation permitted in that dimension when the intended separations are less than standard in the other two. The problem is to choose standards which are safe enough, but not so large as to lead to unnecessary traffic delays and deviations.

1. Introduction. An air traffic control system maintains adequate separation between aircraft by requiring some aircraft to deviate from the paths which they would otherwise follow. In the absence of such deviations it must be assumed that civil transport aircraft would choose to operate at the most economical combination of airspeed, track and altitude (within the limits set by inaccuracies of weather prediction); A.T.C. deviations therefore involve economic penalty to the operator. It is important to assess the magnitude of such penalties for two reasons:

(a) To provide a yardstick by which to compare different A.T.C. systems.

(b) To give a measure of how much might justifiably be spent on new systems and equipment, to enable separation standards (and hence the penalty of deviations) to be reduced without reduction in safety.

In 1964 during the second single handed transatlantic race Colonel H. G. Hasler in Jester lost the correct time and, having no radio, could not get a time check. Thus he could not obtain the longitude but, being a fine seaman and practical navigator, made a landfall on Nantucket light-vessel by a method used by old sailing ship skippers. This was to position himself on the correct parallel of latitude and sail due east or west to the destination.

The International Regulations for Preventing Collisions at Sea, 1960, became effective on 1 September 1965 and, in spite of the sound advice contained in the Annex, it already seems doubtful whether the Steering and Sailing Rules will prove efficacious in preventing collisions. It is clear that a growing body of opinion considers that changes of some sort are necessary; the April number of this Journal, for instance, contains interesting proposals by Captain J. F. Kemp.

Captain García-Frías has made the following comments on the discussion of his recent paper (this Journal, 18, 141) contributed by E. S. Calvert and S. H. Hollingdale.

Captain García-Frías observes that while Calvert and Hollingdale recognize that this paper combines his earlier suggestion for a sector rule with a set of manœuvring rules, they have failed to recognize the fundamental distinction between the structural and operational aspects of an encounter; his proposed manœuvring rules deal with the operational aspect. These rules differ from the manœuvres they have themselves proposed in that they depend only on the bearing of the target as seen on the radar at a distance of 2 miles; they derive from his sector rule and do not require a quantitative estimate of the miss distance. His rules are only the same as their manœuvres for the case of a steady bearing since their manœuvres also require the sign of the miss distance.

Many discussions have taken place regarding the feasibility of plotting when using radar in reduced visibility and it soon becomes apparent that there is confusion in what each of us considers by the term ‘plotting’.

These notes are to suggest definitions of the various stages of plotting and to consider the conditions under which each is necessary.

Radar plotting can be divided into three stages:

(a) Record plotting

(b) Deductive plotting

(c) Forecast plotting.

In Captain Wylie's note on ‘Radar as an anti-collision aid: the ultimate essentials in presentation’, masterly definition by an acknowledged expert aptly crowns the discussion which follows Commandant Oudet's paper on ‘Collisions and the Courts’ in the same issue of the Journal. It is interesting and gratifying to find in that discussion a suggestion by E. S. Calvert on manœuvres in close-quarter situations which recalls a similar proposal made by the French Institute at our first joint meeting held in London in 1957. It is that the manœuvre should be a turn of 2x° to starboard if a threat, an echo for example, is perceived x° forward of the port beam. This suggestion passed almost unnoticed at the time although it is mathematically sound if one assumes that the speed of the two vessels is the same.

In many practical situations in the field of navigation, it has been noticed that the probability distribution of measured errors has a shape which has considerable departures from that of the normal distribution. These departures are particularly noticeable in the ‘tails’ of the distribution of practical cases, which are often higher than those of the corresponding normal distributions (see, for example, ‘Is the gaussian distribution normal’, W/Cdr. E. W. Anderson. This Journal, 18, 65).