ELECTRONIC AUTO-ROUTING
FIELD-TESTING A WORK-IN-PROGRESS TECH
Safety Measures with a New Tech
Car navigation has never been easier since the invention of satnav systems. Just feed in the zip code and the system knows where you are and where you want to go, and presents a suitable route. Now, some navigation apps such as Waze are even crowdsourcing data so that the nav can suggest alternate, faster routes in the event of roadway accidents or work slowdowns. Why not do the same with our chart plotters? Out at sea, there are no defined tracks except in channels and harbor entrances, making your A-to-B route more of a free-form adventure. The depth of water can be restrictive, and there can be other no-go areas on or under water that may limit the chosen route so the computer cannot work its magic.
Garmin, pioneers of road navigation, introduced their autorouting system for chart plotters many years ago. Like the landbased system, you simply entered start and finish points on the chart plotter, and the system would find the route automatically. Now, electronic chart provider Navionics has developed an autorouting system that does much the same thing. The Garmin system is limited to use on their own branded plotters, of course, but Navionics does not make plotters, just electronic charts, so their system has been adopted by many of the electronic players, such as Raymarine and the Navico Group, which includes Lowrance, Simrad, and B & G. With the competition heating up, it is time to look at what these automated systems can and cannot do, including analyzing the risks of letting the electronics take over your passage planning.
Having your route planned automatically would be fine if the seas were not littered with obstructions to navigation. These days, cruisers have to cope with traffic separation schemes, oil rigs and platforms, and now the first wind farm in U.S. waters in Long Island Sound. In addition, there are the normal problems of avoiding rocks, shoals, and shallow waters. Navigation has become complicated and the question asked was, “Could auto-routing cope with all of these challenges while still providing a safe course?”
SAFE MARGINS
Both of the systems tested allow you to input the draft of your yacht as well as other parameters, such as mast height (for bridge clearances), beam (for canals and locks), and tolerances for how close to the dangers you want to pass. As for draft, you don’t want to input your exact draft because the system takes this figure literally: give yourself plenty of wiggle room to be sure to avoid unnecessary groundings. Mast height and beam only apply to inland waters and canals, but the distance off dangers seems to be a bit arbitrary. This only applies with the Garmin system and is defined by words like close and distant. If you are going to define this parameter, then it would be helpful to have it as a measurement, such as ¼-mile. With auto-routing, it is up to you to define the safety margins and you have to remember that these two systems do not take into account the rise and fall of tides when considering safe water depths.
HAZARDS
Unless dangers to navigation are marked on the electronic
chart, auto-routing will ignore them. There could be areas of strong tidal currents that you might want to bypass, but you are unlikely to see these on electronic charts. The increasing navigation restrictions posed by offshore oil installations and wind farms will only be taken into account if they are displayed on the chart, which requires you to have the latest charts installed. Offshore oil installations and wind farms make life difficult for many European boaters, and are steadily increasing the number of no-go cruising areas.
ROUTE PLANNING
Let’s look at how the two systems cope with plotting specific routes. This review is not a direct comparison between the two systems but a look at how the systems cope with a variety of challenging routes. We were limited by the charts available for the two systems, which were the Northeast and Northwest for Navionics and Florida and the Bahamas for Garmin. In selecting the plotted routes, we tried to provide a variety of challenges to auto-routing, including sailing from major ports, crossing traffic separation zones, sailing along shipping channels, dealing with wind farms, and making landfall.
GARMIN
To start, a route was plotted from Miami Beach Marina to a port on Marathon Key. There are three options when making this passage:
1) The inside channels which are littered with shoals and bridges.
The Garmin system chose the last of these three, probably ignoring the inside channel because of the many shallow-water patches. We entered 10 feet as a safe draft, and this might have been the reason that the inside/outside channel was rejected. If I had been navigating, I think I would have chosen the second option that somewhat avoids the strong current that pours up the Straits of Florida when you go outside.
The proposed course made a dogleg out of the marina and headed down the middle of the channel coming out of Miami Harbor. Then it made a sharp right turn to pass inside the spoil areas. After that it plotted a route that I would have considered dangerous, passing inside the buoys that mark the various shoal patches that exist along this shore. In some cases the routes passed close to shoals where there was no marker buoy, such as Pickles Reef. The plotted route passed near to these shoal patches, much too close for comfort, as far as I was concerned. To follow this route you would be totally dependent on the accuracy of your GPS, with no margin for error. Around Key Largo, the proposed route took a more inshore line between shoal patches, and then toward Marathon, the routing took the buoyed channel west of Pigeon Key Banks before turning east toward the marinas at Marathon.
To be sure, it was a tough route to plot, so the next one we asked it to configure was more straightforward: head from Miami across to Nassau, Bahamas, the route followed by many boaters. This is a fairly straight shot across the Florida Strait from Miami, and then Garmin’s auto-routing software recommended a voyage around the northern end of North Bimini, following a track to pass close to the north side of the unmarked Moselle Bank rather than taking a more inside route to pass the marker on North Bank. From there, it suggested crossing the Flats, passing close to the North West Shoal and then on to Nassau. In Nassau Harbour, the route skirted the cruise-ship terminal and the shallow water in order to arrive at the marina. Apart from that channel north of North Bimini, this was a pretty safe route to follow as long as you had functioning GPS to ensure safe passing of some of the shallows.
Summing up, you would need to do a lot of fine-tuning to make sure you had a viable route with the Garmin system, and you would need to check out the route in considerable detail to ensure safe passing distances for many dangers.
NAVIONICS
For Navionics, we threw in some areas with traffic separation zones which have strictly governed rules about what you should
and should not do in these areas according to the Colregs. You have to cross separation channels at right angles and if you are under 20 metres in length, you should try to keep out of them altogether. Over that length, you have to use the channels, but because Navionics does not request your boat’s length, the system does not know. In fact, it ignores these routes, except to flag up a warning triangle to show that you should be aware. The Garmin system also ignores these traffic zones.
The route from Seattle to Victoria on Vancouver Island is littered with traffic separation zones, but the system completely ignored them and took the straight-line, shortest route available. Although the plotted route displayed a multitude of cautionary triangles, I would have opted for a route closer inshore, away from shipping lanes and only crossed them at right angles on a couple of occasions rather than zigzagging in and out of them. This auto-plotted route would have required a lot of adjustment and quite frankly it would have been a lot easier just to start from scratch, and plot the whole thing manually.
A second route was plotted from Montauk, Long Island, to Sunset Lake on the northeastern corner of Martha’s Vineyard. This route, once again, ignored traffic separation zones heading into Rhode Island (apart from the warning triangles). There was also one point where the route chose to pass inside a channel marker buoy with no advantage in the distance covered. Although relevant to the discussion (but not to this example), the new wind farm to the south of Block Island was displayed on the chart.
COMMENTS
These trials showed that auto-routing has a way to go before it provides a completely trustworthy system of passage planning. On relatively simple routes where there are no or very few restrictions, it works reasonably well, but the algorithm used is not geared to coping with traffic separation zones and has trouble dealing with challenging routes. Both systems ignore separation zones, and unless you correct the course manually, you are subject to arrest by the Coast Guard for violating the Colregs. It is also evident that any proposed route relies heavily on having accurate GPS positioning because the clearance from dangers can be quite small. Our route down the Keys passed much too closely to dangers for my liking. I plotted an additional route in European waters that took me right across a shoal where the chart did show enough water, but that shoal was noted for its constantly shifting depths.
Both systems will offer you a route into and out of harbor, directly into your marina berth but quite why you should need auto-routing in buoyed channels I am not sure. In buoyed channels it is pretty clear where you need to be and auto-routing does not keep you to the starboard side of the channel. In open seas, auto-routing might have some merit by offering a basic route around which to start planning.
The display makes it quite clear that the proposed routes are not to be used for navigation without detailed checking and double-checking. You do not want to follow it implicity.
n
These trials showed that auto-routing has a ways to go before it provides a completely trustworthy system for passage planning.