Remember the good old days when in-car navigation meant asking for directions or, if you were being a bit flash, buying a map of an unfamiliar city from a petrol station en route? Seems like a different century now doesn’t it, now we have become so familiar with modern satellite navigation systems?
Over 14,000,000 people, or half the UK’s driving population, use a sat-nav on a regular basis, but have you ever stopped to think about how it works – and how it might be used in the future?
How your sat nav works
Your sat-nav (or iPhone/Android app, they all work in the same way) detects the signal from a number of satellites – Galileo will be the dominant European system, with 30 satellites planned to be in orbit by 2020 - and triangulates your position from that.
Think about that, for a moment. You sat-nav calculates your position by triangulating the speed-of-light signal from three or more satellites in the blink of an eye.
And it continues to do so continually, adjusting your position in real-time while simultaneously showing your speed and direction of travel to a greater degree of accuracy than would have been possible even with NSAS-style technology only a few years ago. It truly is a miracle of modern science.
How accurate is my sat nav?
The accuracy of your sat-nav depends on how many satellites it can see. With six satellites in view, it’ll be accurate to within about 20 metres but if it can see 10-12 then that accuracy rises to within five metres.
How a sat nav calculates a route
The calculation of a route is possibly its most impressive feature. I mean, imagine how many infinitely variable routes there must be between my home in North Wales and the Saga offices in Kent!
So, how does my sat-nav calculate the optimum route between those two points?
Well, it starts by imagining an ‘as-the-crow-flies’ route before breaking that route down into stages uses the nearest roads as a guide. It then divides that road network into a series of lines (or vectors) between fixed points. These fixed points are generally road junctions, as junctions usually offer a binary left/right choice, making the subsequent calculations relatively straightforward.
So, to get to the Saga office the first part of my route (the first vector) must be along my drive; that’s a singularity. But at the bottom of my drive I could turn either left or right, a binary decision and one that would get me to the main road no matter which one I chose. In sat-nav language, I have two vectors to choose from.
The way the sat-nav decides which to recommend is to attribute two values to every vector: length and cost.
In my fictional commute, one direction is a bit longer than the other but the traffic is generally free flowing (i.e. this vector has a long length but a low cost), while the other road is shorter but more prone to school-run congestion (a short length but a high cost). So, depending on what parameters I’ve set (a fast route or a direct route, for example) my sat-nav will calculate a value for both vectors, selecting the one that best balances length and cost.
It continues to do this at every vector of my journey until it has a suggested route, created by linking thousands of decisions into a coherent road route.
It doesn’t stop there though, because most sat-navs now offer a choice of two or three different routes, so it has to do the whole computation two or three times. This is something to reflect on next time you’re feeling impatient as you wait for it to calculate your route!
Five optional extras that will add value when you come to sell your car
Almost all sat-nav systems can now adjust your route to take into account the prevailing traffic conditions, diverting you if there is a jam ahead. Again, it will assign a value to the delay and calculate whether the optimum route is to go another way or whether you are still better toughing it out on the original road.
This can get a bit clunky, with the first-generation of sat-navs sometimes taking a detour to save a couple of minutes, but they’re getting better and fuzzy logic is helping to calculate a genuine ‘human’ saving as opposed to one that is purely mathematical.
Waze is a great example of a system that uses crowd-sourced data to calculate the optimum route depending on user-reported obstructions, traffic jams and other incidents. I’ve used it extensively now and it has never let me down, failed to detour me around heavy traffic or otherwise improved my sat-nav experience. (It also reports the location of fixed and mobile speed traps and the location of police officers and vehicles…)
By linking sat-nav systems together in the future, civil engineers will be able to direct cars and other vehicles along multiple routes, depending on their size and the priority assigned to them. Lorries, for example, might be directed along a route that avoids city centres and low bridges while smaller, more nimble vehicles will go along a different, more direct, route.
Enjoyed this article? Why not sign up for our Technology and Motoring newsletter?
Linking your sat nav to your car
Evolution will come as a result of very clever people thinking of all the different ways we can utilise all that beautiful information your in-car sat-nav gives us. As ever, the possibilities are only limited by the imagination of the engineers working on these projects, and cross-fertilisation between non-automotive engineers (like those at Apple and Google) will help enormously: car manufacturers are notoriously staid in their thinking and they’ve only begun to scratch the surface of what’s possible.
For now I’ll illustrate the sort of things you can look forward to by telling you about two very different manufacturers: Jeep and Rolls-Royce.
The GPS signal in the Rolls-Royce Wraith gives the eight-speed automatic gearbox advance warning of an impending hill or corner, enabling it to pre-emptively change down into a more responsive gear. The result is Jeeves-like anticipation and a seamless gearchange that happens in a supernaturally unobtrusive fashion. It is very clever and genuinely useful.
Jeep, on the other hand, coming from a culture of consumer litigation and risk-avoidance, uses exactly the same information in a very different way. It uses the sat-nav data to predict the severity of upcoming corners, cross-referencing that to the speed of your vehicle. If the Jeep’s electronic brain thinks you are going too fast, it shrieks hysterically at you to slow down. This sounds like a good idea but that same culture of consumer litigation and risk avoidance means that the threshold has been set too low, leading to too many false positives and an overwhelming urge to turn it off. The result is frustrating in the extreme and is so annoying that it ruined the car for me.
Let’s just hope they get it sorted before they decide to link it into the braking system, eh?
Are road maps making a comeback?
Using your sat-nav system in a built-up area can lead to the signal bouncing off buildings and walls, reducing its accuracy. In a worst-case-scenario, this might mean the signal is ‘off’ by hundreds of metres.
A sat-nav system can also throw the odd hissy fit, sending you in completely the wrong direction. The most extreme example might be the Syrian lorry driver who was diverted through Skegness as he made his way from Turkey to Gibraltar, but we’re all familiar with stories of people driving into lakes because they slavishly followed their sat-nav’s directions. Remember, your sat-nav is an aid, not a substitute for commonsense!
Eyes on the road…
It is estimated that driver distraction is a factor in up to a quarter of all road traffic accidents, and research accrued out in Sweden showed that using a sat-nav is as distracting as using your mobile phone.
So, it’s important to pull over where it is safe to do so before programming or otherwise making adjustments to your sat-nav.