Feature request : Display line of sight

[rooart]

🚀 feature request : Display altitude and line of sight

Context

I've been trying to switch from Locus Map to OsmAnd+ for hiking. I have found many great advantages with OsmAnd including live updates, search feature, POI display, custom map rendering and so much more. But there are still 2 missing features that force me to open Locus Map during my hikes. I have the feeling that they would be quite simple and quick to implement in OsmAnd so I am suggesting them here.

Description

• A text label showing the exact altitude/elevation under the cursor (center of the map). => Today you can only display your current elevation as a text field. You cannot display the elevation of any point by simply exploring the map. The only way to know the elevation of any point is to zoom in and estimate the elevation from surrounding contour lines. Let's call it the "old fashion way" of reading a map. But when you have DEM files available, why not use them to simply display the elevation why flying over a point with your cursor ?
• A line of sight showing the direction you are looking at / heading to. => Want to know the name of the summit you are looking at ? Why not use the compass to display a line on the map and easily recognise the points of interest around you ? Some will say that there is already a feature to orientate the map according to the compass. Yes it works for POIs that are close to you. But when trying to distinguish a summit amongst others which are several kms away from you, you are missing the precision of a line drawn on the screen.

Illustrated Proposed solution

[vshcherb]

We already have an open issue to display altitude at each point

[rooart]

We already have an open issue to display altitude at each point

Oops, my bad: I search for "elevation" before posting and found nothing, but you are right I can see 2 related issues for altitude display:

• 11265

• 10194

The need is the same: the proposed implementation is different. With the first one you need to click on the map to get the information => additional action, less fluid from where i stand but implementing both would not hurt... With the second one (coordinates widget), you still get X and Y coordinates on top of the altitude, which is more heavy on the display if you do not need them. Honestly I would prefer my proposed implementation. But anyway 3 open issues showing there is a real need for it 👍

And what about line of sight ?

[vshcherb]

Line of sight is something new and we're wondering whether small simple triangle doesn't help enough. We're not sure whether most of the phones could provide any similar precision to have as a line.

[rooart]

Line of sight is something new and we're wondering whether small simple triangle doesn't help enough. We're not sure whether most of the phones could provide any similar precision to have as a line.

Line or triangles can work but from experience (looking at summits) the difficulty comes when you need to zoom in on the map at some distance from you. Meaning that the map is not centered anymore on your location, but is centered on the object you are trying to identify, and your location is out of the map. This is why you need an "infinite" display such as a line (or a line with triangles as per picture).

[jandr]

The proposed "line of sight" surely can be useful and interesting (although as mentioned, I am not sure how the compass and gps would be to be able to use it in practice).

However, another use for the "line of sight" which I would find interesting is if it would be possible to draw a line between two selected points on the map, and determine whether there is an unobstructed line of sight between them. My use case for this would be when planning photography sessions or when setting up antennas for communication, which require unobstructed line of sight.

Thanks!

[fedcas]

Line of sight is something new and we're wondering whether small simple triangle doesn't help enough. We're not sure whether most of the phones could provide any similar precision to have as a line.

Line or triangles can work but from experience (looking at summits) the difficulty comes when you need to zoom in on the map at some distance from you. Meaning that the map is not centered anymore on your location, but is centered on the object you are trying to identify, and your location is out of the map. This is why you need an "infinite" display such as a line (or a line with triangles as per picture).

I second this line of sight thing. Actually there is an app just for that, but it's buggy from my experience. Also, why install a separate app based on OSM when you could just use osmand and its offline maps? It should be really easy to implement since the compass orientation is already there.

[vshcherb]

Point of elevation is implemented, line of sight is not that precise as we want it, so it should be more like a sector than.

[fedcas]

A circular sector would be great. If you are concerned about the precision maybe a popup with a disclaimer and reminder to calibrate the compass could help.

Thanks anyway for the great work!

P.S. i forgot to link the app I was mentioning if you want an example (but it's buggy and not updated), it's called OSM peak finder

[sonora]

@vshcherb Question regarding

line of sight is not that precise as we want it, so it should be more like a sector than.

Why is that? Surely the device orientation is simply a discrete direction, derived from the magnetic sensors? So it's the same value we derive the "Map orientation by compass" from at rest? So we offer to display the map rotated accordingly, why could we not also offer to display that value as a line?

[fedcas]

Random ideas:

• Maybe a circular sector + its bisector axis. Graphical way of telling azimut +- tolerance.
• The sector width could be either fixed or getting wider as estimated compass accuracy gets lower (some apps do that, even google maps for its small blue sector... can't remember about osmand).
• When zooming in places far from the current location, of course the movements of the axis or sector borders would be jerky, as linear measures are angle*distance (in other words the sensivity to small compass rotation will become larger and larger when moving far from the current location). To mitigate that a gradient could be used to fill the sector within its two borders instead of just being an on/off overlay color (with maximum opacity at the axis position). So that if for example you are trying to look for a summit, you zoom in and you "lose" the axis because it goes out of frame, you can easily slowly rotate the device towards the axis line guessing its location from the gradient.
• The gradient may also be applied after some reasonable distance (maybe 100 o 1000 km or so) to make the sector and the axis fade away.

[sonora]

My point is we have no measure of compass accuracy in Android as far as I am aware. There is, certainly, inaccuracy, but it depends on e.g. when you last calibrated by twisting through all axes. I would not know how to measure and quantify the current inaccuracy, hence have no basis for what cone angle to draw.

But would advocate drawing a line, damped by Kalman filtering for usability.