Loss of sensor input
One occurrence, which will activate a warning, is the loss of sensor input such as arises if log or gyro compass data is missing. It is important here to note that the ARPA has no way of knowing what values to expect and so can only warn of their absence.
(The warning ‘log error’ means that the ARPA is receiving no input from the log, and not that the value it is receiving is in error.)
Use Of Raster Scan Marine Radar Displays
The attention of Mariners is drawn to recent research which has highlighted the possibility of misleading or erroneous displays occurring with Raster Scan Displays (RSD) that suffer from the loss of certain input signals. These problems are said to occur in two areas; loss of video input and loss of azimuth signal.
The research showed that in some cases loss of video input resulted in freezing of the picture, an effect not noticed until the range is changed. The cause appears to be related to the fact that the screen image is generated by a video processor and if the signal is lost, the display does not redraw or refresh.
In other instances, the loss of azimuth signal led to rotation of targets or targets being depicted on wrong bearings.
In many cases, the RSD did not display an alarm, or indicate in any way that there was a problem with signal input.
Mariners should investigate the type of RSD fitted to determine the response of their system to loss of input, particularly video and azimuth signals. If no warnings are displayed in these circumstances, then procedures should be developed to periodically test the integrity of the display.
Mariners should contact the manufacturer of the equipment for advice in detecting input failures and guidance in developing test procedures.
The above is particularly important for operators of high speed craft; with the limited response times on these craft meaning that early detection of system faults is imperative.
Further information can be obtained from the research
paper, reprinted below from Focus,
This alarm quantifies departures from the predicted tracks of targets. The target(s) activating the alarm will be indicated.
If all the targets generate the track change alarm then it becomes obvious that the alarms were activated by large or rapid manoeuvres performed by own vessel. In general, this condition can be recognised, as all targets will exhibit the track change symbol.
This alarm is generated to offer automatic warning of own vessel or other vessels dragging in an anchorage. If a known stationary target (for example, a small isolated navigation mark) is acquired and designated as such then an alarm will be activated if the designated target moves more than a preset distance from the marked position. If the stationary target appears to move, then it must be due to the own vessel dragging her anchor.
Alternatively, it will also give a warning if another ‘tracked’ vessel in the anchorage moves away from the anchorage.
Since there is a limit to the number of targets, which an ARPA is capable of tracking, in areas of high traffic density, there may well come a time when all the tracking channels are in use. This is particularly likely when automatic acquisition is in operation. An alarm will warn the operator to inspect the untracked targets for potential dangers and to transfer tracking from less important targets, which are being tracked to the potentially dangerous ones (not as yet tracked).
Wrong or invalid request
Where an operator feeds in incorrect data or data in an unacceptable form, e.g. course 370˚, an alarm and indicator will be activated and will continue until the invalid data is deleted or overwritten.
Time to maneuver
Where a ‘delay’ facility is provided with trial manoeuvre, an alarm may be provided to alert the observer, to the fact that, say, one minute until time to manoeuvre’.
Safe limit vector suppression
This facility, if selected, suppresses the vectors of targets whose predicted motion does not violate the safe limit and is an attempt to reduce ARP ‘clutter’.
The ARPA continues to track the target whose vectors are suppressed. If any of them should manoeuvre in such a way as to violate the set safe limits, the vector of that target will reappear and the safe limit alarm will be activated.
If a decision is taken to use this facility, be aware to switch off the facility before contemplating a manoeuvre.
This facility is the same as the safe limit alarm but operates only when the trial manoeuvre is selected. It is not available on all systems.
An isolated land target (lighthouse with a Racon) with good response is selected as reference. It is acquired and tracked by one of the ARPA tracking channels and then designated as a fixed target. This makes it possible for the tracker to calculate the ground track of own vessel and hence to maintain the movement of the electronic origin of the display in correlation to it.
When using this facility the observer should be particularly watchful for other targets, which approach the reference target, and, in particular, for those which pass between the observing vessel and the reference target. If the target moves too close to the ‘echo ref.’ target chances of target swop may be greatly increased.
In general the same stabilization is applied to the radar picture presentation and to the true vectors, i.e. either both are sea-stabilized or both are ground-stabilized. Thus in general, where automatic ground-stabilization is selected, true vectors will indicate the ground tracks of targets and not their headings.
Failure to appreciate this can render the presentation dangerously misleading if it is mistakenly used in the planning of collision avoidance strategy.
One might expect the danger of observers being misled in this respect to be less than in the case of a raw radar display because, except in case of an along-track tide, there will be angular displacement of own vessel’s vector from the heading marker.
The above makes it possible to have true-motion parallel indexing. It also makes it possible to maintain electronic navigation lines and maps in a fixed position on the screen.
However, it must be stressed that the presentation may not afford traffic heading information and may therefore in principle be unsuitable for collision avoidance.
Automatic ground-stabilization can also be achieved by using the output from a twin axis Doppler log that is locked to the ground or feed from the GPS.
Whenever ARPA is used in the True track mode, data relating to own vessel’s motion is fed in from the speed log and from the gyro/magnetic compass.
Assuming that the speed log is feeding in the vessel’s speed through the water and is not on the ‘bottom lock’ mode, then the displayed true track of the vessel would be sea stabilized.
Vectors would therefore indicate the true track through the water of other vessel’s as well and thus would also the visual aspects of the other vessel’s, irrespective of ant tide/current experienced.
IT IS THEREFORE VERY IMPORTANT THAT WHEN ARPA IS USED IN THE TRUE TRACK ANTI COLLISION MODE, THAT IT IS ONLY USED IN THE SEA STABILIZED MODE.
The above is the reason that in spite of a vessel being equipped with a GPS receiver, it is compelled by regulation to carry an operational speed log. The ARPA has to have a feed from the speed log.
Coastline drift may be prevented by feeding in the set and drift due to the current/tide, or by having the feed come in from the speed log working on ‘ bottom lock’ condition. Or also by incorporating the CMG obtained from the GPS.
Another way is to have the facility of ‘echo reference’ lock on to a stationary target (selection of the same requires utmost care, and is not recommended for the novice).
Under the above the display becomes ground stabilized. The displayed vector will then indicate the targets true tracks. Of course due to the potentially misleading effect of the data relating to the tracked vessel’s aspect, this mode should not be used when assessing collision risk or planning avoidance strategy.
There are advantages of using either a True or a Relative motion display. Relative motion displays and subsequent plotting gives an immediate indication of which ships are on a collision course.
On the other hand, whether a target is stationary or moving can be usually distinguished more readily with a true motion display.
Generally any one of the displays may be used, however with the inherent advantage for collision avoidance, relative motion maybe more suitable for open sea condition for collision avoidance.
Now regarding whether to use Ground stabilization or not.
Well ground stabilization display may and will give a misleading idea about a target/ship in coastal areas, involving tidal currents.
GPS speed in general gives ground speed, and there lies the necessity of having a speed log, which can give input to the Radar of the set and drift experienced by own vessel.
In the following example the same is highlighted:
The above is a case of an own vessel observing another target in an area where the current is a factor. If ground stabilization is used, then the own vessel course is taken by the ARPA as 000 deg. And speed of 12k, however due to the current the actual vector of own vessel is
Thus unless sea stabilisation is used, the plot will give a totally erroneous result and will seem that the vessels are passing clear when actually they would be colliding.
This necessitates the use of a speed log as is mandatory under SOLAS.