# RE: Speed, heading and elevation graphic

From: Thomson, Martin <Martin.Thomson@andrew.com>
Date: Mon, 15 Feb 2010 07:12:53 +0800
To: Steve Block <steveblock@google.com>, Angel Machín <angel.machin@gmail.com>
CC: public-geolocation <public-geolocation@w3.org>
Message-ID: <8B0A9FCBB9832F43971E38010638454F032E566DB2@SISPE7MB1.commscope.com>
```Not that this matters a great deal, but how is this applied to common use cases?

Imagine a car that is climbing a steep hill.  Let's say that it's moving at a speed of 10m/s with respect to the ground in a North-East direction on a 30 degree incline (i.e., _very_ steep).

The other way of describing this is using strictly horizontal speed, as Steve suggests: speed[h]=8.66, heading=45, vspeed=5, but is that really how you would describe this motion?

Given that a car speedometer measures ground speed, rather than horizontal speed, that's arguably an easier set of values to report.  The vup/vdown is no easier or harder for a GPS unit to produce; it's just a matter of determining the dot product (and inverse cosine thereof) of different sets of vectors.

To consume, the story might be different.  But certainly, producing a second orientation is easier than a vertical speed.  Arguably, using ground speed is also a better fit with how people think of movement.

--Martin

> -----Original Message-----
> From: public-geolocation-request@w3.org [mailto:public-geolocation-
> request@w3.org] On Behalf Of Steve Block
> Sent: Friday, 12 February 2010 9:57 PM
> To: Angel Machín
> Cc: public-geolocation
> Subject: Re: Speed, heading and elevation graphic
>
> > After reading this publication
> > (http://web.math.hr/~rogina/cartography.pdf) I would suggest using
> > something similar to this nomenclature:
> > - Speed: projection of the velocity vector over the local tangent
> > plane (LTP). Could be decomposed into Vnorth and Veast.
> > - Heading: orientation angle of the speed vector (the same we have in
> > the current version of the spec)
> > - Vdown (or Vup): vertical speed, being positive downwards (or
> upwards).
> Yes, that sounds like a good approach, and probably better matches
> common use cases

```
Received on Sunday, 14 February 2010 23:12:21 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:33:56 UTC