Jump to content

SLOW DOWN


manxmike

Featured Posts

I think on phones and tablets other factors can affect GPS accuracy. For instance on Android (apple might be the same) google use wifi to give a rough instant position while the GPS gets a fix....this initial position isn't very accurate but does refine after a few seconds. I also have a GPS speedometer app and as well as speed, it displays the current positioning accuracy. This is usually about 60ft when it starts and improves down to 6ft once multiple satellites are received. However sometimes it takes a minute before you get to 6ft accuracy, which would affect low speed accuracy initially. 

The other thing to remember when you are using GPS speed (particularly in a car) is that in some receivers it is only dead accurate when you are travelling straight and level. Going around corners and up/down hills affect accuracy GPS. Some receivers can correct for this and some can't.

Link to comment
Share on other sites

2 hours ago, IanD said:

 

A setting like this changes the algorithms that are used to smooth the position and estimate speed from it, so the slow-speed accuracy is better (and less noisy) but it responds to big changes in speed more slowly.

 

Obviously resolution and accuracy are not the same, just because it reads to 0.1mph doesn't stop the reading jumping up and down even at constant speed -- this is exactly the behaviour I was referring to earlier when I mentioned benchmarking. If you plotted the tracks some devices wobbled from side to side like a drunk, others gave a much smoother and more accurate track.

 

There's a massive amount of signal processing going on here to try and get from the raw and noisy location signal (which is updated tens of times per second) to reliable position/speed estimates, Kalman filters and all sorts of predictive tracking for those who are interested. Devices (or apps) which have had more effort put into optimising this can perform much better. There are also differences between how good the GPS antennas are, ones with low gain will give bigger errors and more noise in both position and speed.

 

And almost none of this is properly tested, just "will it give a location and how long does it take to lock" -- which is also down to the quality of the hardware and the algorithms... 😉

 

(I use the GPS Test app too, it can also read down to 0.1km/h (or mph) but I haven't tested it while walking or on a boat)

The first time I had contact with GPS I think we waited something like 8 hrs for a satellite pass with an external aerial and a bit of kit the size of a tea chest. Its was all mumbo jumbo to me, I was just facilitating it, all to see if a platform was where we thought it was.

Link to comment
Share on other sites

4 hours ago, doratheexplorer said:

I do a bit of running now and then.  I was tracking my runs on my iphone (is that GPS?).  I got to be able to do a reasonably comfortable 5km run.  So I went to a ParkRun (which is an organised, weekly 5km run in a park).  I put my phone app on as usual.  I was much dismayed to find that when my app said '5km', I still had a full half km to run!!  I asked one of the organisers about it, and she said the course was accurately measured out and that phone apps frequently overestimate distance.  Interestingly, I discovered that I could zoom right into the maps of my run, and when I did, it looked like I'd been drunkenly zig-zagging from side-to-side all the way.  Except I knew I hadn't.  Then I bought a gps watch, which gave much more accurate readings.

 

It was as if the gps in my phone was constantly trying to pinpoint me and failing. 

 

I have a phone that can sit on my coffee table and record 50km of travel in a week .... very handy if you want to play Pokemon Go.  Yesterday it thought I was in Cropthorne and not Cheltenham.

Link to comment
Share on other sites

7 hours ago, Hudds Lad said:

Don't forget there's also perceived or mental speed, if the stationary boat does not hear a change in engine note they assume you're speeding.

I tend to slow well before passing, but leave a little bit on so i can then knock that off when in earshot ;) 

This effect gets very interesting when you have a hydraulic drive, as we do. If you knock back the engagement of the swash plate then the fluid is pumped at lower pressure so the propeller (and hence the boat) slows down. However, the engine is then working less hard so the revs actually go up. It gets quite funny watching people come leaping out of the cabin red-faced and ready to shout about slowing down, only to realise we are crawling past. I once got filmed by an angry boater on the S&W as we went past at near full revs but less than 0.5mph!

 

Alec

Edited by agg221
Link to comment
Share on other sites

On 04/05/2022 at 07:45, manxmike said:

Given that I haven't been on the canals for over five years I think "constantly" is a bit ott. 

If you have never experienced boats going past at speed, causing your boat to rock then I can only wonder how you moor up. Boats travelling too fast, causing breaking wakes have been a fact of life since the canals were created. There were boats towed by galloping horses in the 18th century - they were known as packet boats and were equipped with cutters to sever the tow lines of slower boats that got in the way. I wonder if they caused other boats to rock and roll.

Long may you continue to live in a world of peace and tranquility. 

 

The "cutters" were there for show, nothing else.

Think about it.

 

Link to comment
Share on other sites

21 hours ago, Hudds Lad said:

Don't forget there's also perceived or mental speed, if the stationary boat does not hear a change in engine note they assume you're speeding.

I tend to slow well before passing, but leave a little bit on so i can then knock that off when in earshot ;) 

and there are those who just shout at every passing boat. I have had that experience when passing a boat a boat length after exiting a lock! Even if I had wanted to, there was little chance of speeding.

  • Greenie 1
Link to comment
Share on other sites

16 hours ago, IanD said:

 

A setting like this changes the algorithms that are used to smooth the position and estimate speed from it, so the slow-speed accuracy is better (and less noisy) but it responds to big changes in speed more slowly.

 

Obviously resolution and accuracy are not the same, just because it reads to 0.1mph doesn't stop the reading jumping up and down even at constant speed -- this is exactly the behaviour I was referring to earlier when I mentioned benchmarking. If you plotted the tracks some devices wobbled from side to side like a drunk, others gave a much smoother and more accurate track.

 

There's a massive amount of signal processing going on here to try and get from the raw and noisy location signal (which is updated tens of times per second) to reliable position/speed estimates, Kalman filters and all sorts of predictive tracking for those who are interested. Devices (or apps) which have had more effort put into optimising this can perform much better. There are also differences between how good the GPS antennas are, ones with low gain will give bigger errors and more noise in both position and speed.

 

And almost none of this is properly tested, just "will it give a location and how long does it take to lock" -- which is also down to the quality of the hardware and the algorithms... 😉

 

(I use the GPS Test app too, it can also read down to 0.1km/h (or mph) but I haven't tested it while walking or on a boat)

It may read  to 0.1 but it is not accurate to that precision. When developing an application that used raw GPS data I spent some time looking at the data stream. I experimented with several algorithms to find one that was suitable for the context ie a canal boat. The basic problem is that the speed in the data stream varies even more than my 1960 Ford speedo used to do. There will no doubt be some processing within the GPS unit itself (it has to to derive speed from location) so this reflects the fact that the location data is not all that accurate anyway. In my case the readings span at least +- 0.5 mph when cruising at what seems to be a constant speed. So, the need is for a suitable smoothing algorithm. The challenge is to balance between the competing needs of responsiveness (how quickly do you want the read out to show a change) and stability (how much noise in the reading can you tolerate?). I settled for stability so when slowing down for moored boats I only really see the change of speed when I am passed - unless at Golden Nook, that is (:))

 

Link to comment
Share on other sites

8 minutes ago, Mike Todd said:

and there are those who just shout at every passing boat. I have had that experience when passing a boat a boat length after exiting a lock! Even if I had wanted to, there was little chance of speeding.

 

I had the same on the River Soar in Leicester, at the point where it's maybe 100' wide and I was perhaps 60' away from the moored boats. No I didn't slow down...

Link to comment
Share on other sites

I have always found the variation in displayed speed on a GPS, when the throttle has not been touched,  coincides with the variation in depth/ width of the canal.

  • Greenie 1
Link to comment
Share on other sites

10 minutes ago, Mike Todd said:

It may read  to 0.1 but it is not accurate to that precision. When developing an application that used raw GPS data I spent some time looking at the data stream. I experimented with several algorithms to find one that was suitable for the context ie a canal boat. The basic problem is that the speed in the data stream varies even more than my 1960 Ford speedo used to do. There will no doubt be some processing within the GPS unit itself (it has to to derive speed from location) so this reflects the fact that the location data is not all that accurate anyway. In my case the readings span at least +- 0.5 mph when cruising at what seems to be a constant speed. So, the need is for a suitable smoothing algorithm. The challenge is to balance between the competing needs of responsiveness (how quickly do you want the read out to show a change) and stability (how much noise in the reading can you tolerate?). I settled for stability so when slowing down for moored boats I only really see the change of speed when I am passed - unless at Golden Nook, that is (:))

 

All of which is what I said -- with correct algorithms (not just averaging, but adaptive predictive filtering -- effectively, modelling the physics of the moving object) you can get reliable and accurate speed measurements down to 0.1mph, certainly at low speeds. But then you don't get fast response to rapid speed changes (acceleration/deceleration) -- the simplest fix is to have manual slow/fast modes as mentioned earlier, or you run both algorithms in parallel and switch between them when the speed is changing, or the most sophisticated one is to make the algorithm coefficients vary dynamically depending on what had been detected, and then you can even try and work out what kind of "vehicle" is in use (e.g. foot, bike, car, boat) and change the algorithm adaptation to suit-- which is what we did in the GPS chipset.

 

What is done nowadays I have no idea about because this is all just hidden inside the phone SoC -- maybe they do the cleverest version, maybe the dumbest one, who knows...

Edited by IanD
Link to comment
Share on other sites

The 'best' walking GPS system have both WAAS and EGNOS which (apparently) makes the accuracy far better, but depends where in the world you are :

 

The Wide Area Augmentation System (WAAS) is an air navigation aid developed by the Federal Aviation Administration to augment the Global Positioning System (GPS), with the goal of improving its accuracy, integrity, and availability.

 

 

The European Geostationary Navigation Overlay Service (EGNOS) is the first pan-European satellite navigation system. It augments the US GPS satellite navigation system and makes it suitable for safety critical applications such as flying aircraft or navigating ships through narrow channels.

 

Accuracy. An EGNOS-enabled receiver provides location accuracy to within 3 metres. Without EGNOS, a standard GPS receiver only provides accuracy to 17 metres.

 

When looking at a track of my walks it actually shows which side of the road you have walked on and even when you have left the road for a vehicle to pass you.

 

Link to comment
Share on other sites

19 hours ago, ditchcrawler said:

My Ipad is no where near as good as my cheap Redmi phone

screen shot from Ipad track

image.png.6d967a182384c3c36cf23bc1c76cdc77.png

 

Screen shot from phone track

image.png.81570c265ca4f70d804f011a006c56fd.png

 

Also as expected the distance travelled is different, further on the Ipad as it counts all the deviations

Link to comment
Share on other sites

3 minutes ago, ditchcrawler said:

screen shot from Ipad track

image.png.6d967a182384c3c36cf23bc1c76cdc77.png

 

Screen shot from phone track

image.png.81570c265ca4f70d804f011a006c56fd.png

 

Also as expected the distance travelled is different, further on the Ipad as it counts all the deviations

Looking at the iPad track I see your steering is a good as it's always been 🤭

  • Haha 1
Link to comment
Share on other sites

22 hours ago, Loddon said:

I own at least three devices that have GPS all made by different manufacturers, strangely they all read the same so its highly unlikely that they are not correct. One of which I use to maximise my speed on the road and if that was wrong I would have got a ticket by now.

If they were  all the same make then there is a remote possibility that the manufacturer has got it wrong but they are not.

 

In any case its far more likely to be correct than some bod on the towpath.

But being accurate at driving speeds is not the same as being accurate at boating or walking speed. I have a walking GPS app on my phone which I sometimes use to measure boat speed, but the displayed speed varies quite a lot, presumably as individual gps satellites come into and go out of view through the trees, buildings etc. What it does show though is just how much the boat slows down going through bridges and other narrows even though the engine speed has not changed.

  • Greenie 2
Link to comment
Share on other sites

2 hours ago, IanD said:

 

I had the same on the River Soar in Leicester, at the point where it's maybe 100' wide and I was perhaps 60' away from the moored boats. No I didn't slow down...

I got told to slow down by the rowing club there!

And at Holme on the Trent..

  • Haha 1
Link to comment
Share on other sites

On 04/05/2022 at 15:34, Idle Days said:

It seems to me to be something like country villages where everyone demand that you slow down when driving through their village.

Yes, I was doing a genuine 32mph in a 30 zone the other day, only to be overtaken by a 4 x 4 which slowed after 100 yards or so to turn into a house drive.

  • Sad 1
Link to comment
Share on other sites

Another problem with GPS on the canals is the effect of "urban canyons" you get in cuttings or lots of tree cover where the unit loses signal, same as if in large cities with towering buildings.

 

I once tried to plot our days journey on the Shroppie using my Garmin Etrex 30, which uses WAAS, EGNOS & GLONASS for maximum sat coverage.

As we'd set off from somewhere near Tyrley heading for the Staffs & Worcs you can imagine the result :D 

Apparently we'd hit some cracking speeds between where it lost signal and then re-acquired it due to some averaging it must have worked out.

Until they invent dedicated personal GPS sats that can follow you directly overhead like a drone, results may be spotty in certain terrains. ;) 

  • Greenie 1
Link to comment
Share on other sites

18 minutes ago, Hudds Lad said:

Another problem with GPS on the canals is the effect of "urban canyons" you get in cuttings or lots of tree cover where the unit loses signal, same as if in large cities with towering buildings.

 

I once tried to plot our days journey on the Shroppie using my Garmin Etrex 30, which uses WAAS, EGNOS & GLONASS for maximum sat coverage.

As we'd set off from somewhere near Tyrley heading for the Staffs & Worcs you can imagine the result :D 

Apparently we'd hit some cracking speeds between where it lost signal and then re-acquired it due to some averaging it must have worked out.

Until they invent dedicated personal GPS sats that can follow you directly overhead like a drone, results may be spotty in certain terrains. ;) 

It's a miracle that GPS works at all given how tiny the signal is -- even in good conditions it's invisible below the receiver noise floor, and in poor ones it's much weaker still, it's only thanks to miracles of signal processing that it gets picked up at all... ,-)

Link to comment
Share on other sites

Just now, IanD said:

It's a miracle that GPS works at all given how tiny the signal is -- even in good conditions it's invisible below the receiver noise floor, and in poor ones it's much weaker still, it's only thanks to miracles of signal processing that it gets picked up at all... ,-)

They have improved massively, I remember the early units were basically useless under a leaf and considering my job is quite often in and around trees...

the modern units very rarely lose the signal, still a bit useless for plotting accurately in dense woods but I can forgive them for that

Link to comment
Share on other sites

Just now, tree monkey said:

They have improved massively, I remember the early units were basically useless under a leaf and considering my job is quite often in and around trees...

the modern units very rarely lose the signal, still a bit useless for plotting accurately in dense woods but I can forgive them for that

Better signal processing and algorithms nowadays, also more satellites (not just GPS any more)... 😉

 

In many cases the phone already knows roughly where it is which greatly speeds up initial locking, some units even download ephemeris data (satellite positions) via the Internet. If you know where the satellites are and roughly where you are locking is much faster.

Link to comment
Share on other sites

1 hour ago, Hudds Lad said:

Another problem with GPS on the canals is the effect of "urban canyons" you get in cuttings or lots of tree cover where the unit loses signal, same as if in large cities with towering buildings.

No such thing as urban canyons it tree cover where I boat these days.

It's good being in the land of the "Big Skys"

Especially where the speed limit is 7mph 😉

  • Happy 1
Link to comment
Share on other sites

23 hours ago, David Mack said:

But being accurate at driving speeds is not the same as being accurate at boating or walking speed. I have a walking GPS app on my phone which I sometimes use to measure boat speed, but the displayed speed varies quite a lot, presumably as individual gps satellites come into and go out of view through the trees, buildings etc. What it does show though is just how much the boat slows down going through bridges and other narrows even though the engine speed has not changed.

Not entirely correct: car speeds are much greater so the same absolute error is a much smaller percentage at 70 mph than at 3 mph. An error that would be well within the ticket-collecting bounds will be a large variation for a canal boat. The accuracy calculations strat from a location accuracy - which is what is actually measured (well, almost!) and speed is a derivative.

Link to comment
Share on other sites

3 minutes ago, Mike Todd said:

Not entirely correct: car speeds are much greater so the same absolute error is a much smaller percentage at 70 mph than at 3 mph. An error that would be well within the ticket-collecting bounds will be a large variation for a canal boat. The accuracy calculations strat from a location accuracy - which is what is actually measured (well, almost!) and speed is a derivative.

Indeed, but ticket matters apart, knowing your speed in a car is 70 mph +/- 3mph is fine. Being told your boat is doing 3mph +/- 3 mph is useless!

Link to comment
Share on other sites

4 minutes ago, David Mack said:

Indeed, but ticket matters apart, knowing your speed in a car is 70 mph +/- 3mph is fine. Being told your boat is doing 3mph +/- 3 mph is useless!

 

But even this is not what happens. You get told your boat is doing 3mph +/- <a deadly secret value they never tell you>

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.