You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Just like speed and weight should be separated, it might be good to separate turn time and turn weight.
In my experience (with bike routing), people prefer a route with few turns, even if it's slightly longer, since It makes the route easier to remember and to follow. As width speeds/weights, if you only have the time penalty to tweak and not a separate weight, there's no way to achieve this without distorting the travel time.
If additional data becomes available to the turn function, you might want to avoid turning to/from/across specific types of roads or lanes based on this data, without affecting the travel time.
@emiltin from my understanding, this should fall out of #77 if implemented correctly. We might need to adjust the turn function to return both a time and a penalty, but apart from that we should be fine. Timing-penalties need to be added to the time, weight penalties to the weight part and the rest should be handled transparently.
Just like speed and weight should be separated, it might be good to separate turn time and turn weight.
In my experience (with bike routing), people prefer a route with few turns, even if it's slightly longer, since It makes the route easier to remember and to follow. As width speeds/weights, if you only have the time penalty to tweak and not a separate weight, there's no way to achieve this without distorting the travel time.
If additional data becomes available to the turn function, you might want to avoid turning to/from/across specific types of roads or lanes based on this data, without affecting the travel time.
Related to #2918.
The text was updated successfully, but these errors were encountered: