Tag Archives: navigation

SatNav of the future?

Galileo satellite in orbit.

I always love it when the more theoretical aspects of physics, so often the exclusive realm of mathematicians and those studying the origins of the universe, have a direct impact on our daily lives. 

Take the GPS system, for example, which has so many challenges that it really shouldn’t work at all, as far as I can see.   Quite apart from the very clever software and signal-processing needed to allow a weak radio signal, from a satellite tens of thousands of miles above you, to be diluted over a substantial chunk of the earth’s surface, and distorted by the ionosphere, and compromised by the weather, yet still be picked up by a small battery-powered device in your pocket when you’re sitting inside a moving metal box surrounded by lots of other noisy electrical signals… quite apart from all that, there’s another problem… and it involves Einstein.  

Did you know you were using his Theories of Relativity whenever your satnav tells you how far it is to the next junction?    I’ve always found this very pleasing.

GPS depends on being able to make very accurate time measurements, and the issue is that the satellites are travelling very fast, which means that time itself runs at a different rate for them than it does for the those of us down below driving round the M25. Their atomic clocks run more slowly than an equivalent on Earth.  But a bigger effect comes from the fact they’re in a lower gravitational field, which causes them to run faster!  Pleasingly, this means that both Special and General Relativity need to be taken into account by the GPS system, in order to work out these differences and stop a gradual drift, which would be visible over time on your satnav map and might make you think you were on the wrong road.  Thanks for sorting that out, Albert!

However, for all the magic of GPS, it has its limitations, as you’ll know if you’re driving through a long tunnel, or between skyscrapers on the streets of Manhattan.  And the delicacy of the GPS signal also means that it can easily be interfered with by, for example, an enemy on a battlefield. When the GPS signal is lost, navigation systems have to fall back on ready-reckoning using accelerometers and distance measurements to try and guess where you are.  This is a process as old as navigation itself, but it is very fallible, because your knowledge of your current position is based on your position a little while before, and any errors in that process get magnified the further you go. The accuracy of accelerometers has greatly improved over time and become less dependent on things like the bearings of spinning gyroscopes, but it’s still a problem.  

This is why a lot of people are awfully excited about the recent experiments with Quantum Positioning, which may offer a much more accurate way to do this in the future (and I would guess may be useful long before quantum computing!)  Once again, theoretical physics may help us find out where we are.  It may be rather a long time before you have this mounted on your dashboard, but perhaps not so long until it’s in ships, submarines and aircraft.

So how do you use quantum mechanics to work out your position?  (Or, more precisely, your acceleration, from which changes in position can be derived?)

Fortunately, there’s a really excellent video by Ben Miles explaining the basics.  Nicely done.

Google Tip of the Day

Here’s a quick two-and-a-half minute video which might save you some time one day, if not now!

Measuring distances and areas in Google satellite view

(A direct link is here, in case you can’t see the embedded video.)

Watching the map

A screenshot from the appI’ve long been a fan of the ViewRanger app for handling maps on my phone. I know I’ve been using it for at least 8 years, because I have tracks recorded from early 2012.

With it, you can use free maps, and buy commercial ones from a range of sources, in many countries, and pay for them in a variety of ways: subscriptions, individual purchases, or — the method I use because I started before the days of in-app purchases — buying a block of credits up front and using them as needed to purchase the map tiles you’re interested in. ViewRanger isn’t perfect: I find bits of its user interface very counter-intuitive, but in general it’s served me very well.

Another worthy contender for UK users, by the way, is the app UK Map, written by a good friend of mine, and which I’ve been using even longer than ViewRanger. UK Map has always provided a great combination for users here: a standard basic UK road atlas, always available offline; the ability to download the free Ordnance Survey maps (since almost the moment they became available), and the overlaying of paths from OpenStreetMap, which are often the most up to date indications of where you can actually walk. More recently, it too gained the option to buy premium maps, but by then I was somewhat invested in the ViewRanger ecosystem. Nonetheless, I do recommend UK Map – it’s helped me out on more than one occasion when I found myself completely out of phone coverage and just needed a good old-fashioned street atlas, for example. Don’t be fooled by its lack of flashy graphics and fewer immediately-obvious features. Phil, the author, is a very smart guy and knows what he’s doing. It’s worth having on your phone.

Maps have always held a fascination for me, and their combination with technology especially so. I remember I was in Seattle on the day the iPad was released, and so was able to get one that day, about 6 weeks before most of my European contemporaries. (I’m still a bit smug about that!) It only took me a few hours, though, to realise it was the best map-viewer I had ever seen. A zoomable, infinitely-scrollable, tactile display that was also big enough to show a reasonable area? Amazing. (The main thing I miss, by the way, about no longer also having an iPad Mini is that it combined most of that with the ability to fit in a coat pocket.)

Anyway, this is all old hat now. We almost all wander the footpaths and byways gazing at our phones, rather than unfolding bits of paper from our map pockets. They do, after all, have that undeniably useful extra feature – the little blinking dot or crosshairs showing where you actually are. And zooming and scrolling mean that the place you’re heading for is no longer on the other side of the fold! I do find, though, that zoomability means I’ve lost the intuition I used to have about how long it would take me to walk a certain number of inches…

There’s another problem, too, with phone-based navigation, that I’ve found in the soggier parts of the UK: touchscreens, and indeed fingerprint readers, really don’t work very well in the rain. The process of extracting the phone from a damp pocket, trying to find something less damp to wipe it on, failing to unlock it with my fingerprint and having to type a PIN before trying to manipulate a semi-responsive touchscreen has just occasionally left me thinking that even damp paper would be preferable to electronics.

Until today.

Because today I discovered something I should have noticed some time ago. ViewRanger has an interface for the Apple Watch. If it’s running on my phone, I can simply raise my wrist to see where I am.

I could have used this in the past, had I discovered it, but it’s also vastly improved by my recent purchase of a Series 4 watch to replace my Series 0, which means that when I raise my arm to look at my watch, things actually appear before my arm gets too tired and has to be lowered again. In fact, while testing it on this afternoon’s dog walk, it was effectively instantaneous. Where does this footpath to the right go? Let me just glance at my wrist – ah, OK. You can scroll around, zoom in and out etc, but most of the time, all I needed was a quick glance: no unlocking, no soggy touchscreen. (It also looks amazing, but I think that’s mostly because of another recent purchase: a new and expensive pair of spectacles!)

All of which makes me wonder whether, despite being at the opposite extreme in terms of screen size, I may after a decade have found an even better map viewer, at least in certain circumstances, than the iPad!

What 3 words?

This is a brilliant idea. Take the world and divide it into 3m squares. Then, given a modest vocabulary, you can label each square using just three words.

For example, I work half-time at ‘faced.ears.sport’, which you can find by typing it into the What3Words map at map.what3words.com, or you can turn it into a handy URL:

http://w3w.co/faced.ears.sport

Isn’t that much easier than saying ‘The south west corner of the William Gates building at 15 J J Thomson Avenue, Cambridge, CB3 9JW, UK’? Or ‘52.210577 N 0.092133 E’?

It’s even more valuable, though, in countries where addressing schemes are less well established or non-existent.

Now, it has a couple of limitations that I can see. First, you do need to be fairly precise about those words if, say, you’re reading them over the phone. If, instead of ‘faced.ears.sport’, you went to ‘face.ears.sport’, you’d find yourself in a little residential street in Montana, which would be delightful, but you wouldn’t find me there on a typical work day. ‘faced.ears.port’ is in Louisville, Kentucky. However, the fact that they’re so spread out probably makes such errors less likely to go undetected – this is deliberate.

The second limitation is that this is a commercial operation and not an open standard, which is a pity in some ways, but understandable. It’s free for individuals to use – there’s a free iOS and Android app, for example – and the pricing page contains this assertion:

If we, what3words ltd, are ever unable to maintain the what3words technology or make arrangements for it to be maintained by a third-party (with that third-party being willing to make this same commitment), then we will release our source code into the public domain. We will do this in such a way and with suitable licences and documentation to ensure that any and all users of what3words, whether they are individuals, businesses, charitable organisations, aid agencies, governments or anyone else can continue to rely on the what3words system.

I think it’s a brilliantly simple idea. The concept has been used in other situations (passwords, PIN numbers etc), but works really well here.

Sending satnav locations from your iPhone/iPad to a BMW navigation system

One of those ‘in case you’re Googling for it’ posts! This will probably be of little interest to anyone who doesn’t own both an iOS device and a BMW, but might be useful if you own both.

Also available on Vimeo if preferred.

Time travel

My friend Mike Flynn has been working for some years on very fast routing algorithms — routing as in maps, that is — and his primary demonstration of this is TimeToAnywhere – a system which can work out how long it takes to drive from one location to everywhere else on the map.

So you can say, for example, “There’s been an accident here. Which ambulances could reach it in less than 15 mins?” Or, “I work in Dry Drayton. Where could I live, and still have less than a half-hour commute?”

timetoanywhere

Each coloured boundary represents 10 minutes’ driving.

This is pretty, but those of you with a computing background may also realise that, using most of the standard algorithms, this is also a very time-consuming problem when you try do it across this number of points. Mike, however, measures the time taken by his system in microseconds.

He’s recently set up a demo server which, if it doesn’t get too swamped, is fun to play with to get a feel for the speed! You can find it at TimeToAnywhere.com, and if you want to know how to get the most out of it, watch Mike’s short video.

© Copyright Quentin Stafford-Fraser