Website for optimal EV-routes

[evRoutes]

Hi!

We are a small group of developers and EV-enthusiasts who made a website for EV-drivers to find optimal routes between locations. The application is at this stage based on official data of locations for CHAdeMO fast charging stations. Our goal is to provide the best routes possible for your electric car given two end-points, regardless of what car you drive, and based on actual real world distances. To compare with most current route-planners or charge-point finders, we don't simply show you the traditional route for fossil-fueled cars, along with some charging-stations at a fixed distance from that path. We take into account the specific range for each car and use advanced algorithms to choose shortest overall trip:

http://www.evRoutes.com

Please provide us any feedback (in this thread or via the website) which you feel could help us take it from where we are now to something truly useful and beneficial for the EV-community as a whole.

Happy Routing!
The evRoutes team.

 

[Turnover]

I just tried it from Madras, Oregon to Biggs Junction, Oregon. I think this is doable on Highway 97 as it is basically downhill. EV-routes gave me an uphill route on Highway 26 to Mt. Hood then downhill on Highway 35 to Hood River then East on I84 to Biggs Junction. My guess is that route wont work due to the elevation losses and gains between Madras DCQC and the Government Camp (Mt. Hood) DCQC. Elevation changes are a big factor out here in the West. I haven't seen an app that includes that factor longitudinally yet. In fact my Highway 97 "downhill" route - longitudinally- has some ups and downs to it, but I think they are workable with our Leaf. Plan to try it this summer now that the Madras DCQC has opened up.

 

[DNAinaGoodWay]

Love the concept. Hope you can fine tune it and do apps for iPhone and Android.

Tried Rochdale MA to Brattleboro VT. Route was good, but pushing my range in summer, forget about winter. No EVSE on that route, but some nearby, it gave me no alternatives like some map apps.

How will it account for DCQCs that are limited by dealers to business hours or other restrictions?

It asked if my car was a LEAF or a iMiev, but no others. Can it account for my degradation? Ambient air temp? Elevation changes? Speed limits?

Tried Rochdale to Rockport MA. Route was all highway, 86 miles. There are new DCQCs along this route, and I would need them, but again, no mention of them, or secondary road alternatives.

Also, couldn't switch it from kms to miles.

But, I've been waiting for an app like this, I hope you perfect it. Good luck.

 

[GIBBER]

The App should show the location of the DC machines along the suggested route and allow user to click on them to gain hours of operation, etcetera. Keep at it!

 

[DaveinOlyWA]

the ability to network in the status of charging stations like; in use, available, broken, last session, etc would make this app invaluable

 

[DNAinaGoodWay]

It's saying a LEAF's range is 140 km, 87 miles, and while that is possible under certain conditions, it's going to be less with degradation, high speed, hills, cold, snow, etc.

Your using official CHAdeMO locations, but their database is far behind PlugShare, which is mostly what we use in North America. Being crowd sourced, it's the most up to date.

 

[kubel]

I like the intended concept, but it doesn't appear your routing algorithm considers factors such as speed and elevation when producing routes (I noticed it had me take highways twice on a trip from my parents house to my house). I also think you should switch your station database to plugshare, as that seems to be the most complete and accurate resource for charging stations in existence. Finally, consideration must be made to miles/kWh when taking that route with factors considered including time of year (winter, for instance, reduces my range by around 50%), speed, elevation changes, etc...

 

[OakLeaf]

Great idea, and I would also repeat what others have already said - Plugshare has much better data to plan with.

Tried a few test routes from home that would require a stop to charge and they all return an error stating that "There is no charge station close to the start position".

A suggestion would be to also find a way to factor in current and forecast local weather conditions (maybe just temperature is practical). For example, cold weather has a large impact on range due to the battery being cold and the use of heat.

 

[asemeco]

Plugshare.com has a "Trip Planner" option where you specify start, destination, and search radius (how far off-route you are willing to go for an EVSE). Then you can add stations to the route. Finally, you can export the route to Google Maps, although that part seems to be broken for now.
You can get ideas from this:
For example, it would be great if you could choose route options (like Avoid Highways) separately for each segment of the route.
Elevation (uphills) should be taken into account.
When you are driving at 60 MPH on a highway with uphills, it makes a big difference whether you have two lanes available or more. Most people drive at > 70 MPH, and they tend to get annoyed.
You can provide a better integration with mobile devices. I wouldn't mind planning the trip in advance and then 'exporting' it to my mobile device.
I think it is great you are making this effort. I will use it for sure.

 

[Wennfred]

Is it possible to add a Mile button?

Fred

 

[WetEV]

I entered a route from Tacoma, WA, USA to Leavenworth, WA, USA.

First segment that evRoutes gives is from Tacoma to the charger at Sultan. This segment is 109 km. Maybe on a warm day, with a tail wind, with no climate control, newer Leaf. There is, however, little margin.

Using Greenrace ( http://www.jurassictest.ch/GR/" onclick="window.open(this.href);return false; ), and speed of 100km/h, the predicted end charge is 0.2%.

I suggest that you put in these items:

1) Put in a weather/date option: today or stated weather or climatology for some future date.
2) Put in a margin factor. 100% to turtle isn't to everyone's liking, to say the least. I'd suggest a default of 80% starting, charge to 70% at stops, and stop to charge before 30%, all settable by the user, who would surely have different ideas.
3) Put in a capacity factor. 100% new leaf, to 70% almost ready for replacement battery, user settable. Default should be 90%, again to stay near the "no worries" end.

Already suggested are:

4) Include elevation change. The loss from regeneration should be speed and slope dependent. A very steep hill requires friction braking. A mild incline will require power to maintain speed. In the first case, loss is high. In the second case, loss is very low.

5) Include speed, at minimum user entered by segment. Other route options would be nice as well.

6) If you need to charge more than the user setting to make the next leg with the user set margin, this needs to be flagged.

At least, these are my rarely humble opinions on this subject.

 

[LakeLeaf]

Several routes I tried were along plotted along the fastest roads with the more elevation gain instead of the slower, flatter roads.

 

[evRoutes]

Hi again and thank you all so much for these comments - highly useful feedback!

@Turnover: you are fully correct that the algorithm does not take elevation into account yet. This is something we have had in mind for a long time and we will put high on priority now.

@DNAinaGoodWay: Thanks for your appreciation and good summary of things we need to include. Apps for iPhone and Android will be awesome, it is a planned feature from day one and when the web-app is more fleshed-out we will look into migrating to native apps. As you mention, our current displayed route is in theoretically good weather conditions, wintertime the range should be much less. Together with elevation we have set up a temperature as the next things to take into account, we probably look into temperature fist since it seems to be the biggest factor - and easiest to fix.
DCQC's that are limited by dealer to certain hours we can not yet take into account, we will have to strike a deal with station-owners network by network to access that kind of data, alternatively we will look into building a community-based way of gathering that info.
Currently we have only LEAF and iMiev/similar cars as option, since their only current parameter is range (and connector-type) we will soon try to put up more choices and/or a custom range input box or similar.
The DCQC's you are missing might not be in CHAdeMO's official data or they have not yet updated with it, There recently was a new version of this data that we still had not updated to before release.
Secondary route options is nothing we have had in mind for quite some time but we might re-evaluate that considering the feedback we just got. A miles/km switch should not be a biggie to fix.

@Gibber: as mentioned above, secondary route options and/or adding stations to suggested route is something we might re-prioritize, thanks you!

@DaveinOlyWA: Status of charge-stations is definitely in our wish-list, and a network of community-members could very well be the solution to this that we will go for, thank you for mentioning this!

@DNAinaGoodWay (second post): true, the range we specify is not considering degradation, high speed, hills, cold and snow. I'm not sure exactly how to model the impact of that last one but the others should be seen as factors we not yet accounted for, but it is on our roadmap.
Thanks a lot for bringing up PlugShare, its been one of all the databases we had under our radar but from the recent feedback we had from these post I can safely say it should be one of our top priorities, if not the top priority to include/switch to this database for the US charging-locations.

@kubel: thanks for stressing the importance of weather/elevation and bringing our attention plugShare their data seems like the best choice for your location.

@Oakleaf: yes tapping into weather forecast of temperature and factoring in that to our estimated range is a good suggestion and probably the way we will go forward. Again, PlugShare seems awesome.

@asemeco: Thank you for your appreciation! we will look into PlugShare, the avoid-highway-option will be under our consideration and as mentioned, elevation.

@:Wennfred: switching to miles seems to be an easy-to-implement but common requested feature, thank you!

@:WetEV: thanks a lot of the good overview of many of the requests we seen here today, your opinion seem to reflect most others priorities, thanks for nice rule-of-thumb suggestions of those factors you mention.

@:LakeLeaf: thanks for mentioning this, suggesting flat and slow roads by default or as an option may be implemented some time in the future.

 

[DNAinaGoodWay]

We all keep PlugShare updated, so if you could do a deal with them, it would save you from agreements with each and every dealer. I think the new iMiev will have Plugshare onboard their nav screen, so they may be keen to license.

ChargePoint, unfortunately, stopped providing real time data to PlugShare, and their own website is cumbersome. If you could get real time availability data from them, and other QC networks, that would be very helpful.

Have you seen the "Leaf Energy" app? It incorporates a lot of range effecting variables, based on Tony Williams range charts. Its only downfall is the clunkyness of the route planner. You could learn a lot from them.