A review of the Trip Planning Tool

[RiverBrick]

Every trip planning system I've used has problems other than Google Maps...

The other systems probably weren't hyped to end range anxiety, so it is reasonable to expect more from 6.2. Hopefully, the OPs constructive comments will lead to the algorithm being improved.

In addition to assuming you'll observe the posted speed limit, the system may also ignore headwinds for its pre-departure calculations. Only once you're on the road and consuming energy at an alarming rate does it adjust. At least that is my experience with 6.1 and wind and bitter cold.

To be honest, to allay range anxiety for less experienced EV drivers, I'm surprised a greater margin of error wasn't built-in.

 

[Cottonwood]

I am amazed that Elon had press conference for this software feature that has so many real world problems. Tesla calls it "Beta," but I am not sure that is at the "Alpha" phase yet.

See A review of the Trip Planning Tool for further issues.

Mods: consider merging these threads.

 

[Zythryn]

The posted speed limit is 75 in AZ and 70 or 65 in CA. I don't think I would have made it over the pass to Tejon even if I had kept to the speed limit in the slow lane.
There is a "lightning bolt" symbol in the upper right corner and I clicked this but didn't notice that it did anything but I didn't know what it was supposed to do. I don't know why they can't just always show superchargers on the map all the time. There should be no reason to hide them any time. Once it plans a route, it hides the superchargers that are not stops.

Quite possibly, just wanted to check if the speed was a contributing factor or not, sounds like it was.

When I hit the Lightning bolt, it shows me destination chargers, chargers previously used, and superchargers. Superchargers shown in red, all the others in gray.
You can then select any one of them and you will be routed there.

I do agree that it would be nice for the superchargers that are not on your route to still show on the map, while in trip mode, perhaps simply grayed out.

Until Tesla allows you to plug in your speed, times at superchargers and ranges will be based on posted speed limits. If you travel faster, you will need to adjust accordingly.

 

[NuclearPowered]

The posted speed limit is 75 in AZ and 70 or 65 in CA. I don't think I would have made it over the pass to Tejon even if I had kept to the speed limit in the slow lane.
There is a "lightning bolt" symbol in the upper right corner and I clicked this but didn't notice that it did anything but I didn't know what it was supposed to do. I don't know why they can't just always show superchargers on the map all the time. There should be no reason to hide them any time. Once it plans a route, it hides the superchargers that are not stops.

I can add input to this. We charged beyond the required amount per trip planner in Quartzite, drove the speed limit, 75 in AZ and 70 in CA and made it to Indio with 11% remaining. If we had left when trip planner suggested, we would have been forced to slow down or have run out. Now, it was in the low 50s, so maybe it wasnt taking that into account.

 

[taurusking]

I am amazed that Elon had press conference for this software feature that has so many real world problems. Tesla calls it "Beta," but I am not sure that is at the "Alpha" phase yet.

See A review of the Trip Planning Tool for further issues.

Mods: consider merging these threads.

I have not received the update yet but I am not a big fan of Trip planning tool if it does not include way points. Plus reading about real world problems make me think so.

Also this whole 6.2 update for cars without auto pilot hardware is not much useful at all.

 

[jbcarioca]

So far all the EV trip planners seem grossly inadequate. I am accustomed to aviation trip planners, which routinely take into account temperatures, expected winds and other conditions, not to mention that running out of fuel in the air is slightly more inconvenient than doing so in a Tesla. Either way, I expect round trip planning, incorporation of expected temperatures, winds and elevation changes as well as road repairs, typical road congestion and road conditions. All that information is easily available and the data to infer realistic range expectations and optimal charging procedures certainly is not a huge trick. So, what has nobody done better? Elon was a bit over-the-top and a little ridiculous suggesting that the trivial improvements he mentioned actually make trip planner a useful thing. I suggest Tesla go talk with any number of Tesla owners with experience managing and/or designing air transport trip planners. They can make mincemeat of this relatively simple problem.

 

[COrental]

What happened to destination chargers in this routing software, I thought they were supposed to be included in the options. i tried a route from Denver to Santa Fe a normal 380 mile route. There is no supercharger along the way yet, but a HPWC for a 4-5 hours would maybe do the trick. Instead it routed me through Moab and added 600 miles of scenic Detour onto this otherwise short trip, disappointed with this version, hopefully we will see some quick improvements.

 

[jerry33]

What happened to destination chargers in this routing software, I thought they were supposed to be included in the options. i tried a route from Denver to Santa Fe a normal 380 mile route. There is no supercharger along the way yet, but a HPWC for a 4-5 hours would maybe do the trick. Instead it routed me through Moab and added 600 miles of scenic Detour onto this otherwise short trip, disappointed with this version, hopefully we will see some quick improvements.

As far as I know, there is no trip planner that can optimize a route using both SC and non-SC points for shortest distance combined with shortest charging time. The main thing that would be required for this would be a field in which you entered either the average distance per day you want to travel (where it would then suggest locations to stop for the night), or to put in your overnight stops because the charging done overnight takes zero driving time.

 

[arijaycomet]

What happened to destination chargers in this routing software, I thought they were supposed to be included in the options. i tried a route from Denver to Santa Fe a normal 380 mile route. There is no supercharger along the way yet, but a HPWC for a 4-5 hours would maybe do the trick. Instead it routed me through Moab and added 600 miles of scenic Detour onto this otherwise short trip, disappointed with this version, hopefully we will see some quick improvements.

I do not think those chargers are part of the Trip Planner, but they are part of the range anxiety fix. Meaning, if you are running low on power, it should include those chargers (or at least ones you've used before) to notify you that you're starting to get too far away from the nearest known charger. Trip Planner won't route you through them, but 6.2 will use them as safety nets to make sure you don't go too far.

 

[andrewket]

I found another bug last night. Driving from Williamsport, VA to NoVA, trip planner suggested using both the Glen Allen SC and Woodbridge SC. My home is about 110 miles from Glen Allen. Ok, perhaps it's trying to optimize charge time, but this by itself is a bit silly. But that's not the bug.

I chose to ignore this plan, and instead have dinner at Glen Allen. I charged to 90% (in fact the car was done before our food arrived). When I returned to the car, I clicked ok on the new "you now have enough energy to continue on your trip" pop-up. The nav system no longer showed Woodbridge on the trip window as a destination, but the nav system was still giving me instructions to Woodbridge. So it seems the trip planner realized stopping at Woodbridge was not necessary, but the nav system didn't update the destination to my home. To get it to clear, I had to cancel the trip and reset the destination to home, at which point it correctly navigated directly home skipping Woodbridge.

Although in reality we ignored the tesla nav and used waze. Traffic was bad because of Easter and there was a major accident on I95. Portions were a complete standstill. Waze re-routed us via US1. The tesla nav kept trying to route us back to I95 behind the accident.

Until Tesla gives us 1/ the ability to adjust the assumptions (speed, SOC% threshold, and maybe a preference between optimize charge time vs minimize stops), and 2/ waypoints, I'm afraid the trip planner is not going to be as useful as I was hoping it would be.

 

[Vger]

I certainly hope Tesla is looking at this feedback, and has some mechanism (beyond the early access cars) to record actual vs. expected results from these trip plans. There is obviously a lot of room for improvement, and much of it seems like low-hanging fruit-- problems with easy (to program) solutions, once recognized.

 

[mspohr]

I certainly hope Tesla is looking at this feedback, and has some mechanism (beyond the early access cars) to record actual vs. expected results from these trip plans. There is obviously a lot of room for improvement, and much of it seems like low-hanging fruit-- problems with easy (to program) solutions, once recognized.

I'm sure they are working to improve the software.
I think now the one feature which would make the biggest improvement would be to allow the user to make changes to the itinerary. Right now, you have to accept their routing in toto and you can't make any changes. If you could add or remove individual stops to the route to fit your schedule (i.e. a longer stop at lunch or a better route), it would be a great improvement.

 

[andrewket]

A quick follow-up to my previous feedback: I'm sure Tesla will continue to improve the software. I recognize that they needed to draw a line somewhere on the functionality that would constitute a release (minimum viable product). 6.2 is a good start. Hopefully they get the feedback contained in this thread; I know they read the forum. If there was a way to provide direct feedback other than blind emails, I would do so. I find the "bug report" feature in the car to be inadequate. It only works for me 50% of the time, and you have to leave a very short message, which is frequently not enough to accurately and adequately describe the situation.

Actually - here is a feature request. Rather than report the bug via the voice "report a bug" feature, it should only open a ticket and capture the current state of the car. In the back-end, Tesla sends an email to your registered email address with a link that takes you to a website where you can provide detailed information. This way Tesla gets the time stamp, all the debug information from the car, and a detailed description from the owner. The owner isn't rushed, and doesn't have to try to cram all the relevant bits into a 8 second voice message. Imho, that would be a much better implementation.

 

[Odenator]

Is there any way to turn off the trip planner feature? When I route from Portland to Olympia, it ALWAYS makes stop at the Centralia superchargers, even if I know I will have more 15-20% left on my battery to reach home. It won't show me the total distance from my start point to my destination.

 

[eye.surgeon]

Remember the mission of the 6.2 trip planner as stated by Musk is to eliminate Range Anxiety, which it does quite effectively even now in beta form. The aren't claiming it's the fastest from A-B.

 

[mspohr]

Remember the mission of the 6.2 trip planner as stated by Musk is to eliminate Range Anxiety, which it does quite effectively even now in beta form. The aren't claiming it's the fastest from A-B.

I think they've gone overboard on the "eliminate range anxiety" algorithm.
Yesterday I drove from home to San Francisco. I checked EVTripPlanner and decided that I would stop at Vacaville SC (121 miles) which would be absolutely no range anxiety with my 200+ miles car charge. I plugged Vacaville SC into the Nav system and it promptly told me that I would need to stop at Truckee SC (15 miles), Rocklin SC (70 miles) and Roseville SC (85 miles) in order to make it to Vacaville SC. Three additional stops! This would certainly eliminate range anxiety but at the cost of three additional stops.
As I was driving and decided (foolishly, according to the Nav) to skip the Truckee SC and was 5 miles up Donner Pass, it demanded that I exit the freeway and return to Truckee to charge. I finally just cancelled the Nav since it was just nattering at me. We drove to Vacaville (with plenty of reserve) and charged there.
I still think the best and easiest thing they could do to fix the Trip Planner would be to give you a way to add and remove stops from the route. This is similar to the way evtripplanner.com works. Currently there is not flexibility in the Trip Planner. You have to take their route and stops in total.

 

[Barlow]

I live on the Isle of Texas where until just recently escape to the other 47 states was impossible. When 6.2 was installed on my MS, I wanted to see if and how Trip Planner would route me to my cousin's in Atlanta. (At the time the Mobile AL SC was under construction.) The suggested route was to drive north to SW Minnesota, across into Wisconsin, down to Chicago and onward to Atlanta. A 830 mile direct trip was now 2800 miles. Tree sides of a square. I'll wait.

But I did notice an item that needs to be fixes besides Trip Planning assuming you can charge up at your destination. In the list of all the SC you have to stop at, the info box has the as-the-crow-flies mileage of that SC from your starting point. The mileage is pulled from the SC list in NAV. This if very misleading in a SC trip list. At first, I thought each mileage was how far alone in your trip you are. Clearly not for the above route from Houston to Atlanta. But on a fairly direct trip that mileage may be taken as the trip's mileage by a driver --- which it not, it's less.

I hope the next update of Trip Planner has that great-circle mileage is left off. Bad info is worse than no info.

 

[ThosEM]

We did nearly 900 km from Portland ME to Annapolis MD yesterday with stops in W. Hartford CT and Hamilton Township NJ. Two weeks ago we made the reverse trip with three stops because we didn't skip the unnecessary superchargers. Through both of these trips we noticed a consistent behavior of the energy estimate that is a bit puzzling. When we strike out on any given leg, our estimated SOC at trip's end initially drops noticeably and disconcertingly, for example from 20% to 12%. In some cases we got a message telling us to drive slowly to make the destination. Fortunately, this trend ends after 10 minutes or so and reverses, with the estimated ending SOC rising to the initial estimate or above, depending on driving speed. As the estimated margin grows, we of course increased speed in an attempt to use that energy, which is a nice payback for initial conservatism provoked by the initial decline.

Consistently, we also saw an initial consumption rate that was higher than normal, and this appears to be what is driving the estimated margin downward, because as the consumption returns to normal, the original margin estimate does as well. They appear to be closely tied together, because climbing a grade can also cause the estimated margin to decline, or vice versa.

This leaves me speculating that the entire energy estimate is driven by the recent consumption record, which is "normal" when we arrive at a supercharger, but then rises for a while upon leaving the supercharger, eventually returning to normal. My question is then what drives the consumption upward upon departing from a supercharger? Is it that the cooling system continues to cool the battery down after charging at high power? The problem with that explanation is that we also see a similar effect when starting out from our NEMA 14-50 charger that never exceeds 10 kW.

Could it simply be that energy drawn from the battery while the car is stationary is eventually bookkept into the initial consumption for some period when getting moving again?

Any thoughts would be appreciated. This is a puzzling behavior, but also much preferable to the opposite situation in which the projected consumption is underestimated and leads to a persistent deficit that grows with travel time.

 

[simonog]

Any thoughts would be appreciated. This is a puzzling behavior, but also much preferable to the opposite situation in which the projected consumption is underestimated and leads to a persistent deficit that grows with travel time.

I notice that when we leave our 7 KW home charger the first 5 minutes of consumption figures are very high (Wh/m). I have assumed it is a combination of the energy used while the car was at rest plus a larger amount for amending the battery temperature from optimised for charging to optimised for running.