Forum: General Topics
Forums / General Topics / Estimated Time of Arrival
| Subject: | Estimated Time of Arrival | |
ChrisM 13:44 Location: Phone Model: | Hi Stephen, Please could you explain very briefly how the ETA on the Waypoints Text Screen works? I understand that it's based on the distance to the Waypoint and the average speed of travel(and hence the estimated remaining journey time) plus the current time but; -How is the average speed calculated? Say I had to drive 20 miles on minor road, then the last 5 on a motorway, is the average weighted on the current speed, or just the average over the whole of the current track segment? Or something else? -Is the speed of approach to the waypoint calculated taking the angle of travel into account, (ie the speed will be faster if I'm heading directly towards the waypoint than if I'm on a bit of a tangent? -Any other interesting facts about the way it works? I noticed it behaving a bit strangely the other day, although the estimated time to destination was about 15 minutes, the seconds on the ETA time were ticking up as if it were a live clock. Regards, Chris. | |
Stephen 16:13 Location: Phone Model: | Hi Chris, " How is the average speed calculated? " As you guessed, it just uses the average speed from the current track section. In a previous version (before there were separate track sections) it would use a very smoothed average of the current speed, but I found that this would make the ETA fluctuate a bit too much - its only supposed to be a genaral guide after all... " Is the speed of approach to the waypoint calculated taking the angle of travel into account, (ie the speed will be faster if I'm heading directly towards the waypoint than if I'm on a bit of a tangent? " No - its purely calculated via the distance as the crow flies divided by average speed. I did consider various options like you've suggested but couldn't think of a reliable way to do it. As soon as you factor in direction then other problems crop up, eg: - when travelling tangentially to the waypoint (eg moving round the edge of a circle, with the waypoint at the centre), the closing speed would be zero, making the ETA infinite! - when moving away from the waypoint the the ETA would be negative! Ultimately I figure using a Route is the only way to accurately include the bearing in the ETA calculation! " I noticed it behaving a bit strangely the other day, although the estimated time to destination was about 15 minutes, the seconds on the ETA time were ticking up as if it were a live clock. " This is perfectly normal in certain situations. eg You're travelling along at 20mph average then come to a complete stop 10 miles from a waypoint (ie 30 mins away). The average speed will only change once you start moving again (and add more points to the track) so whilst stationary it'll always read 30minutes to destination, and the ETA will just show the current time +30mins. Cheers, Stephen | |
ChrisM 19:25 Location: Phone Model: | " This is perfectly normal in certain situations. eg You're travelling along at 20mph average then come to a complete stop 10 miles from a waypoint (ie 30 mins away). The average speed will only change once you start moving again (and add more points to the track) so whilst stationary it'll always read 30minutes to destination, and the ETA will just show the current time +30mins. " Hmm, I could see those circimstances, but I WAS moving at the time. Maybe slower than the average speed though, so pr'aps that was the problem? " Ultimately I figure using a Route is the only way to accurately include the bearing in the ETA calculation! " But it doesn't do that at the moment...? Not a problem, the current system seems to work reasonably well anyway. Cheers, Chris. | |
Stephen 23:00 Location: Phone Model: | Wow, post number 1000! " Hmm, I could see those circimstances, but I WAS moving at the time. Maybe slower than the average speed though, so pr'aps that was the problem? " Yeah, that would give a similar result, especially if you happened to be travelling tangentially at that time too, perhaps?... " But it doesn't do that at the moment...? " No - a route effectively works like a series of waypoints stuck together, so admitedly it still doesn't take into account the direction of travel, but the 'error' is largely minimised when the distance to the next routepoint is a lot smaller than the total route length, if you see what I mean...!? | |
(You must be logged in to post a reply to this thread)