navit-project.org
http://forum.navit-project.org/

Any possibility to manually force an alternative route?
http://forum.navit-project.org/viewtopic.php?f=8&t=520
Page 1 of 1

Author:  Nezmi [ 17 Aug 2014, 14:45 ]
Post subject:  Any possibility to manually force an alternative route?

Imagine you encounter en route a traffic jam, floods, procession, parade, etc.

It would be really helpful, if you could instruct navit to find an alternative route avoiding some segment of the previous route -- e.g. from here the next 2km or from here to this point ...
(this is about manually temporary entering blocked roads not automatic!)

Have I overlooked such a feature or is it missing?

If the feature is missing, please consider voting on the poll ...

Author:  ZeroOne [ 18 Aug 2014, 12:00 ]
Post subject:  Re: Any possibility to manually force an alternative route?

I tried to vote - but forum says: invalid...

I think the second option is the best because predefined distances can cause a bad rerouting.

F.i. You're on the highway, there's a traffic jam -> you can choose to mark the traffic jam for 3-15km. Let's say 5km.
The accident is right next to the motorway exit. You can't get off the motorway, but you can get on the motorway (usually behind the exit as everybody knows). If you mark 5km as blocked, it's possible or very likely that you block the motorway access too.
So navit will route you off the motorway but not back on the motorway after the traffic jam - maybe another 15km later on the next motorway access.

In fact there should be an option to mark start and endpoint of the blocked route.
As long as there is no TMC option in navit it would be great to make these work manually.

So the workflow would start with choosing the endpoint (on the radio: "Traffic Jam on Highway at Point X - the car line is 3km long - you'll have to wait 30min.") -> click on the map -> choose "Blocked street" -> choose "Street" (maybe you've clicked on an intersection or bridge and there's another street) -> "set blocked length" -> enter "-3" => voila - that's it.

With this option you can also choose not to drive on specific roads by selecting start or endpoint -> choose "Blocked street" -> chosse "Street" -> "Set Endpoint" -> congrats, you've blocked a specific street.

What do you think?

Author:  Nezmi [ 14 Sep 2014, 12:07 ]
Post subject:  Re: Any possibility to manually force an alternative route?

ZeroOne wrote:
I tried to vote - but forum says: invalid...

Alas voting doesn't work for me either ... (I wanted not to be the first voter, but tried after your problem report.)

ZeroOne wrote:
I think the second option is the best because predefined distances can cause a bad rerouting.

F.i. You're on the highway, there's a traffic jam -> you can choose to mark the traffic jam for 3-15km. Let's say 5km.
The accident is right next to the motorway exit. You can't get off the motorway, but you can get on the motorway (usually behind the exit as everybody knows). If you mark 5km as blocked, it's possible or very likely that you block the motorway access too.
So navit will route you off the motorway but not back on the motorway after the traffic jam - maybe another 15km later on the next motorway access.

You are completely right!

ZeroOne wrote:
In fact there should be an option to mark start and endpoint of the blocked route.
As long as there is no TMC option in navit it would be great to make these work manually.

So the workflow would start with choosing the endpoint (on the radio: "Traffic Jam on Highway at Point X - the car line is 3km long - you'll have to wait 30min.") -> click on the map -> choose "Blocked street" -> choose "Street" (maybe you've clicked on an intersection or bridge and there's another street) -> "set blocked length" -> enter "-3" => voila - that's it.

That is a good idea -- iff (=if and only if) the announcement has that format. In Germany typical announcements are of the form "Traffic jam on X between exits A and B [length/delay of Y]." Further the announced length might be outdated, and the actual length meanwhile much longer (or shorter). Of course this can be easily integrated into your workflow.

Further I think it would be useful to enable your workflow not only from the map, but also from the list of routing directions: In the routing directions select a road, then the exits or an exit and a length, and block it.


Generally, we both have very different approaches to reach the same ends -- IMHO both having equal merits in slightly different usage scenarios and personal usage patterns:

* Your approach is to give very concise start and end points.
+ Optimal routing decisions.
+ Enables long term diversions and contributing to OpenStreetMap (e.g. this road segment is blocked until November 2015)
+ Lends itself very well to route planning in advance.
- Depends on concise and reliable data sources (traffic radio in my experience typically isn't ...)
- Needs considerably user interaction, which might not be feasible when driving without copilot.

* My approach is more fuzzy, as I consider most data sources unreliable and prefer an adhoc solution usable for a driver e.g. at a red light stop. I prefer to give a "'navit' avoid that area and give me routing alternatives".
+ Minimal (hopefully) user interaction.
+ Considers not only a minimal rerouting but also large-scale ones. Together with experience or eduacated guesses of the driver this might provides some resilience against imprecise announcements and the avoidance behavior of other drivers.
+ Enables alternative route selection in general, also in route planning.
- Suboptimal routing decisions.
- No long-term learning or mapping potential.
(For this end my personal favorite is the last option of the voting list.)


But honestly I think both our approaches are valuable and worth integrating.

Page 1 of 1 All times are UTC
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/