Native Nav estimating when to charge on long road trip

[Demosthenes]

Posted already at least once previously, but thought it was worth mentioning again. Although the nav in 1.2.6 is improved and the plotting EA charging stations is a nice touch for long road trips, relying entirely on its estimates can be potentially problematic. On one 260 mile leg of my drive from San Diego to Vancouver, the nav plotted an EA station along the way and estimated I’d have 18% battery left when I got there. Unfortunately, I only had 3% left by the time I arrived and this was with ACC set to 70, so nothing crazy. For the last 30 miles or so as I was watching my battery tick downwards I chickened out and lowered my ACC to the speed limit of 65. If I had stayed at 70, I probably could’ve made it…but maybe with 1% left instead. A bit too close for comfort when you’re far from home!

Long story short: the nav estimates do not seem to take into account speed, temperature, elevation etc and more importantly they do not auto adjust in real time, only once you go below 50 miles left and then it automatically tries to reroute you to the nearest charging station. But by then, if you’re in a sparse area, there may not be any good chargers near by.

 

[DeaneG]

...Long story short: the nav estimates do not seem to take into account speed, temperature, elevation etc and more importantly they do not auto adjust in real time, only once you go below 50 miles left and then it automatically tries to reroute you to the nearest charging station. But by then, if you’re in a sparse area, there may not be any good chargers near by.

That's too bad. Google Nav in our Volvo EV does a very accurate job estimating SOC at destination, usually within +-2% after using half the battery's capacity on a trip segment. Tesla does this accurately as well, with a handy energy graph showing the car's prediction for the trip against actual. Certainly a critical area for improvement by Lucid.

 

[Demosthenes]

It seems like the nav based its estimate on an assumed 3.4 mi/kWh efficiency rate. The reality was closer to 3. Also, with a 90 kWh charge, the car estimates that it can cover 360 miles meaning it assumes a rate of 4 mi/kWh. While this number is achievable with the GT on 21s, I don’t think its typical at freeway speeds of 70 or greater. Need better range estimates!

 

[Coltonw011]

I had the opposite experience today. I had a 233 mile trip back from Old Station, the NAV estimated I would arrive home with 22% SoC and I called BS as I was approaching some hill climbs with 30% SoC and the exterior temp was 95 and I had AC running. I pulled into my garage with 22% exactly. Now, I didn’t have to charge on my trip home, but the NAV guesstimated the whole way and was correct. Not sure what would cause the difference

 

[HC_79]

I raised this as a bug with Lucid not long after 1.2.6 was released. Told them the SoC needs to do some form of check every 5 to 10 mins and update according.

 

[Demosthenes]

I had the opposite experience today. I had a 233 mile trip back from Old Station, the NAV estimated I would arrive home with 22% SoC and I called BS as I was approaching some hill climbs with 30% SoC and the exterior temp was 95 and I had AC running. I pulled into my garage with 22% exactly. Now, I didn’t have to charge on my trip home, but the NAV guesstimated the whole way and was correct. Not sure what would cause the difference

Perhaps the nav estimates are more based on 19 inch range than 21 inch range

 

[Sandvinsd]

Perhaps the nav estimates are more based on 19 inch range than 21 inch range

No. I have 19’s and I find the range numbers very optimistic. I don’t trust them at all.

 

[Demosthenes]

No. I have 19’s and I find the range numbers very optimistic. I don’t trust them at all.

Interesting. What does your computer estimate is the range when you are at 100% SOC?

 

[Sandvinsd]

Interesting. What does your computer estimate is the range when you are at 100% SOC?

I have not yet been at 100%. Back calculating from the range and % battery remaining , it appears that it would be about 500 miles, perhaps 505 at 100%

 

[Demosthenes]

I have not yet been at 100%. Back calculating from the range and % battery remaining , it appears that it would be about 500 miles, perhaps 505 at 100%

I see. My 100% tops out at around 450-460. Your number and my number would theoretically be appropriate for the 19s and 21s respectively so I guess it’s not that

 

[TCGT]

Posted already at least once previously, but thought it was worth mentioning again. Although the nav in 1.2.6 is improved and the plotting EA charging stations is a nice touch for long road trips, relying entirely on its estimates can be potentially problematic. On one 260 mile leg of my drive from San Diego to Vancouver, the nav plotted an EA station along the way and estimated I’d have 18% battery left when I got there. Unfortunately, I only had 3% left by the time I arrived and this was with ACC set to 70, so nothing crazy. For the last 30 miles or so as I was watching my battery tick downwards I chickened out and lowered my ACC to the speed limit of 65. If I had stayed at 70, I probably could’ve made it…but maybe with 1% left instead. A bit too close for comfort when you’re far from home!

Long story short: the nav estimates do not seem to take into account speed, temperature, elevation etc and more importantly they do not auto adjust in real time, only once you go below 50 miles left and then it automatically tries to reroute you to the nearest charging station. But by then, if you’re in a sparse area, there may not be any good chargers near by.

I'm not yet an owner so my reply is based the members I've read here who mention using ACC is for some reason not as efficient...perhaps that could be a source of the error (the Nav possibly being based on regular hands-on driving)?

 

[Coltonw011]

I seem to be the odd one out here, but I used ACC for my highway driving because I was too lazy to use my foot for 90 miles. Coming down 44 I had regen the whole way (60 miles travelled used 12 miles range), I5 I was on ACC, Hwy 20 and 29 I was operating the pedal. Again, mine was accurate, I too was skeptical until I pulled in my garage.

 

[Demosthenes]

I seem to be the odd one out here, but I used ACC for my highway driving because I was too lazy to use my foot for 90 miles. Coming down 44 I had regen the whole way (60 miles travelled used 12 miles range), I5 I was on ACC, Hwy 20 and 29 I was operating the pedal. Again, mine was accurate, I too was skeptical until I pulled in my garage.

I’ve always been curious about the notion that extended use of ACC can actually be detrimental to range. Seems to go against common dogma. I can’t really see it being true unless:

1. The regen algorithm is somehow different when ACC gets activated
2. The ACC algorithm doesn’t apply regen braking as well as a human operator and/or accelerates the vehicle much quicker than a human operator
3. The constant ‘on’ of an ACC state is a huge battery drain

Some of this may be dependent on follow distance. I’ve noticed anecdotally that if I set my follow distance at 4 bars, the ACC is perfect and extremely smooth. Mostly at 3 bars too. But at 2 or 1 bar follow distance, it seems much jerkier and there have been times I wasn’t actually confident it would stop or slow in time.

 

[DeaneG]

I have the feeling that Google's own AAOS navigation software uses data from previous trips (by other identical cars on the same route) to be so accurate. Could even factor in weather, rain, etc. Crowdsourced accuracy - the law of large numbers - and one reason to go with a cloud provider rather than using only an algorithm in the car for prediction. Same idea as keeping track of traffic delays but not in real time.

 

[hydbob]

I have the feeling that Google's own AAOS navigation software uses data from previous trips (by other identical cars on the same route) to be so accurate. Could even factor in weather, rain, etc. Crowdsourced accuracy - the law of large numbers - and one reason to go with a cloud provider rather than using only an algorithm in the car for prediction. Same idea as keeping track of traffic delays but not in real time.

If that's the case, then Google's accuracy for the Lucid would be way off as well, LOL.