There are two places where landsmen and seamen are likely to come into conflict: (i) in the pubs around Portsmouth Docks; and (ii) in tidal estuaries such as the Thames. There is, however, an intermediate class of person, represented nowadays almost entirely by more or less amateur oarsmen and women, which is that of watermen. I am proud to consider myself a member of all three.

This paper was the second Presidential Address given by Captain Majendie. It was first published in the Journal in 1958 (Vol. 11, p. 1). It is followed by comments from David Hopkin. For reasons of space, Captain Majendie's paper has been slightly abridged, mainly by omitting a section on the provision of monitoring intelligence.

I have chosen the title of this paper ‘The Display and Use of Navigational Intelligence’, as I believe that our major problems in the future are likely to be as much concerned with the presentation and the use of data, as with its provision.

Artificial neural networks offer an alternative strategy for the nonlinear control of unmanned underwater vehicles (UUVS). This paper investigates the use of a multi-layered perceptron (MLP) network in controlling an UUV over a sea-bed profile and compares the use of applying chemotaxis learning to that of the more commonly employed back propagation algorithm. The results show that, for differing sized MLPs, the chemotaxis algorithm produces a successful controller over the sea-bed profile in an improved training time. Also it will be shown that, in the presence of noise and change in vehicle mass, the neural controller out-performed a classical proportional-integral-derivative controller.

This paper was first published in the Journal in 1954 (Vol. 7, p. 138). It is followed by comments from Michael Richey.

At the time (1542) when Jean Rotz presented Henry VIII with a variation compass, and an accompanying treatise on its making and use, the question of magnetic variation was an acute one. It formed part of a wider controversy concerning the general validity of the sea-chart which had been raised among seamen. As to this, Dr Pedro Nunez, writing a treatise in 1537 addressed to his friend and pupil the Infante Dom Luys of Portugal, said that there were skilled pilots who derided the chart and declared it to be ‘a mais falsa causa do mundo’ – the falsest thing in the world. Hastening to its defence, Nunez dealt at length with the navigating errors introduced by ignoring the convergence of the meridians, and showed besides that a rhumb, or line of constant bearing was a spiral curve on the globe.

Many texts stress the need for marine sextants to be held precisely vertical at the instant that the altitude of a heavenly body is measured. Several authors lay particular emphasis on the technique of ‘rocking’ the instrument in a small arc about the horizontal axis to obtain a good sight. Nobody, to the author's knowledge, however, has attempted to quantify the errors involved, so as to compare them with other errors inherent in determining celestial position lines. This paper sets out to address these issues and to pose the question: what level of accuracy of vertical alignment can reasonably be expected during marine sextant work at sea ?

When a heavenly body is brought to tangency with the visible horizon it is particularly important to ensure that the sextant is held in a truly vertical position. To this end the instrument is rocked gently about the horizontal so that the image of the body describes a small arc in the observer's field of vision. As Bruce Bauer points out, tangency with the horizon must be achieved during the process of rocking and not a second or so after rocking has been discontinued. The altitude is recorded for the instant that the body kisses the visible horizon at the lowest point of the rocking arc, as in Fig. 2. The only other visual clue as to whether the sextant is vertical is provided by the right angle made by the vertical edge of the horizon glass mirror with the horizon. There may also be some input from the observer's sense of balance and his hand orientation.

Most of the radar systems used in operating marine vessel traffic management services experience problems, such as track loss and track swap, which may cause confusion to the traffic regulators and lead to potential hazards in the harbour operation. The reason is mainly due to the limited adaptive capabilities of the algorithms used in the detection process. The decision on whether a target is present is usually based on the amplitude information of the returning echoes. Such method has a low efficiency in discriminating between the target and clutter, especially when the signal-to-noise ratio is low. With modern signal processing techniques more information can be extracted from the radar return signals and the tracking parameters of the previous scan. The objectives of this paper are to review the methods which are currently adopted in radar target identification, identify techniques for extracting additional information and consider means of data analysis for deciding the presence of a target. Instead of employing traditional two-state logic, it is suggested that the radar signal should be allocated in terms of threshold levels into fuzzy sets with its membership functions being related to the information extracted and the environment. Additional signal processing techniques are also suggested to explore pattern recognition aspects and discriminate features which are associated with a return signal from those of clutter.

Following Dag Pike's earlier paper on this subject I would like to discuss some points and address the problems of collision avoidance with High Speed Vessels within the framework of the present International Regulations for the Prevention of Collisions at Sea (COLREGS). Generally, a valid case for limited modifications to the COLREGS may be presented, but they may not be as wanting as suggested by Mr Pike.

This paper was first published in Volume 19, p. 88, in 1966. Parts II and III were included in following issues of the Journal. The paper is Crown Copyright and is reproduced with the permission of H.M. Stationery Office. It is followed by comments from Stanley Ratcliffe.

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.

This paper was first published in the Journal in 1969 (Vol. 22, p. 118). It is followed by comments from John Kemp.

The last twenty-one years have seen some very striking advances in our knowledge of how animals can determine their location. In many cases we have learned that they have available a wider range of stimuli than ourselves for recognizing landmarks and for pilotage within their home area. Thus the associated senses of smell and taste are extraordinarily well developed in some species. The ability of the males of certain moths to detect the scent emitted by females at very considerable distances had long been known. More recently the extreme sensitivity, and selectivity, of fish to waterborne odours has led to an understanding of how they locate their home waters. As but one example, eels have shown reactions to concentrations of chemicals as low as 3 × 10−18, equivalent to but two or three molecules within the fish's olfactory sac. In other cases animals have developed sensitivities of which we have little or no conception. Ecolocation is employed by certain birds, by many marine animals and reaches a peak of efficiency in the case of bats. Not only do the latter detect sounds of much higher frequency than ourselves, they also respond to echoes of sounds they emitted but 0·001 seconds earlier. We have little appreciation of the sensations produced by the pressure-receptors in the lateral-line organs of fish.

May I congratulate Chris Perkins and Tony Redfern for their excellent paper in the May issue of the Journal. For many years I have read in the Journal, and on occasions listened to, learned ladies and gentlemen discoursing on the subjects of safe navigation and collision avoidance. Also to what I believe are their misguided wanderings into ‘ship domains’ and assumptions that ships behave as atomic particles. Thus it was very refreshing to read the latest and practical contribution by Chris and Tony. Well done chaps!

When a surface is mapped onto a plane so that the image of a geodesic arc is a straight line on the plane then the mapping is known as a geodesic mapping. It is only possible to perform a geodesic mapping of a surface onto a plane when the surface has constant normal curvature. The normal curvature of a sphere of radius r at all points on the surface is I/r hence it is possible to map the surface of a sphere onto a plane using a geodesic mapping. The geodesic mapping of the surface of a sphere onto a plane is achieved by a gnomonic projection which is the projection of the surface of the sphere from its centre onto a tangent plane. There is no geodesic mapping of the ellipsoid of revolution or the spheroid onto a plane because the ellipsoid of revolution or the spheroid are not surfaces whose curvature is constant at all points. We can, however, still construct a projection of the surface of the ellipsoid from the centre of the body onto a tangent plane and we call this projection a gnomonic projection also.

Navigation is a human activity with the aim being to arrive at a predetermined destination. In order to find the way to the destination, the use of current input from the actual environment while travelling is needed as well as stored and organized knowledge of the local geography. Although the knowledge requirement has been studied extensively in the form of cognitive maps or other spatial representation, few studies deal with how the knowledge is used together with the input from the actual environment while navigating.

I am disturbed by Perkins and Redfern's paper in the May 1996 Journal.

The COLREGS have been developed over many years of tinkering and tuning. They are not designed only for educated European masters driving big merchant ships. The users; fishermen, yachtsmen, oarsmen, tugmasters, Rhine bargemasters, and Yangtse junkmen, etc, are from all educational standards and from all the world's cultures. The COLREGS are now well-known. There is such a huge investment of time and effort, of learning by millions of different people, that the prospect of tampering with their fundamentals is horrific, even if it would suit a small class of user belonging to the more advanced countries. Change is painful, and too much change, too fast, kills. Compare the practical decision not to change the side of the road on which we British drive: it might be convenient, but it would cost too much. The same applies to the COLREGS.

This paper was first published in the Journal in 1963 (Vol. 16, p. 9). It is followed by comments from Brian Kendal.

The Transit satellite navigation system, scheduled to become operational in the near future, has been publicized sufficiently for its main features to be rather well known. The excellent article by Dr. R. B. Kershner and Dr. R. R. Newton in the April 1962 issue of this Journal provides a particularly thorough description of the system. In the same issue, Dr. H. C. Freiesleben gives a clear summary of the principal alternative solutions to the navigation satellite problem.

I am pleased to see that, in light of my critical evaluation, Arne B. Molander has reevaluated a number of points in his analysis, and has revised his proposals of Columbus's alleged use of the Moon for determining his longitude. However, it is disappointing that these latest proposals again do not seem to have been thoroughly considered.

Since it is clear that Columbus's East–West position fixes are not correlated with visible lunar-planetary conjunctions, Mr Molander's latest hypothesis is that Columbus observed a dark-horizon moonrise or moonset near each conjunction – which usually can be done whether or not the conjunction itself is visible. However, it is entirely unclear how such an observation can be converted into a longitude. Reading attentively, we can glean a few details of this process: only a single observation is required, since some positions are computed within a few hours of the purported observation; and Columbus apparently used the Ephemerides of Johann Miiller, since errors in that ephemeris are alleged to have caused errors in the computed longitude. In his 1992 paper, Mr Molander asserted that timing of the conjunction is not required by this method, because Columbus somehow utilizes the Moon's daily topocentric motion. But how this datum is utilized, and in fact all details of the longitude determination itself, are entirely obscure.

In a paper in the May 1996 issue of this Journal, Ward and Johannessen gave an excellent summary of the work carried our for the General Lighthouse Authorities on the significance of potential interference to GPS. They identified three possible types of interference for their study:

(i) Interference from other systems on board the ship using GPS.

(ii) Interference from emissions generated onboard other ships or at structures in the vicinity of the ship using GPS.

(iii) Deliberate interference from malicious persons.

Mark Dixon suggests (Forum, Vol. 50, 137) that nobody thus far has attempted to quantify the errors from tilt that arise while observing with the marine sextant. The issue in fact, with the related problem of what exactly is the axis about which the sextant is rotated whilst being ‘rocked’ (to define the vertical), was the subject of a lively controversy in the first two volumes of this Journal some fifty years ago. Since the consensus of opinion seems to have been that the maximum error does not necessarily occur at 45 degrees, whereas Dixon's table suggests that it does, some reiteration of the arguments may be in order.

It is not possible from Captain Gyldeń's contribution to fully explain the phenomenon. It is, however, important to note his experience and to encourage other practical navigators to report similar events. It is essential to have the best possible picture of circumstances when GPS suffers problems, before it is decided to close down alternative systems, which might not be affected in the same way.

I believe that it is timely that Dr. B. D. Yallop, Head of HMNAO, has responded to criticisms and the discussions surrounding the inclusion of ‘Sight Reduction Procedures’ and ‘Sight Reduction Tables’ beginning with the 1989 edition of the Nautical Almanac. As my name was mentioned by Dr Yallop, I would like to set out my own views on some aspects of those discussions because I believe that both the ‘theoreticians” and the ‘practitioners” positions have become too polarized.