Using AI for trip planning

[ChrisO]

I haven't seen this suggested before, but I was curious if it would be a viable option.
I have tried both CoPilot and Gemi. I don't really like the interaction with Gemi.
CoPilot seems much more "flowing". As in I don't have to repeat what I have already told it (but note that it "forgets" once you close the web browser).

I was basically using it for a trip that I have done several times. From San Jose, CA to Santa Ana, CA, but before I had a standard range battery and now, I have and extended range battery that gives me more flexibility. But by going with a route that I know I could determine if the AI(s) are giving accurate information.

One important thing is to give it all the details about your Mach-E and ask it to include the weather and elevation changes and criteria like with SOC you are going to start with and what is the min SOC and what you are going to charge to on the road. Once you get it started you can prompt it for different situations you might what to explore. And you might also add in criteria like the ratings of the charger and anything you require at the stop.

I was pretty happy with the results. This goes a long way to fill the gap that Android Auto/SOC charge can't be used when not in the car.

Sponsored

 

[E90alex]

No. “AI” is just a glorified search engine that parses and summarizes results in a pretty format in plain English.

 

[ChrisO]

No. “AI” is just a glorified search engine that parses and summarizes results in a pretty format in plain English.

Yes, and no. What it can do quite well is get the basic information and then do the calculations and such. And it will tell you the assumptions made so that you can correct them if you see mistakes.

For instance, starting with this prompt:
"I have a 2025 Ford Mustang Mach-E Premium RWD with an extended battery. I have a NACS adapter. I want to plan a trip from San Jose, CA to Shinoda Design Center in Santa Ana, CA. I will start the trip with 100% charge, and I don’t want to go below 10% on the trip. At each stop I don’t want to charge to more than 80%. Consider tomorrow's weather, and elevation differences along the route. The 2025 Mach-E Premium with an extended battery has an 88-kWh battery. I will be driving at the speed limit."

Resulted in this:

Great setup, Chris. Let’s break this down methodically and simulate your EV trip from San Jose to the Shinoda Design Center in Santa Ana, factoring in your Mach-E’s specs, charging strategy, weather, and elevation.


Trip Overview

• Route: San Jose, CA → Santa Ana, CA (approx. 370 miles via US-101 S or I-5 S)
• Vehicle: 2025 Mustang Mach-E Premium RWD ER (88 kWh usable battery)
• Start SOC: 100%
• Min SOC buffer: 10%
• Max charge per stop: 80%
• Adapter: NACS (Tesla Supercharger access)
• Speed: Driving at posted speed limits
• Date: Tomorrow (Sept 27, 2025)

Weather & Elevation Considerations

• Weather forecast:
  • San Jose: ~72°F, sunny
  • Central Valley: ~85°F, dry heat
  • Santa Ana: ~78°F, mild breeze
  • No rain or extreme wind expected—ideal for range efficiency
• Elevation profile:
  • San Jose: ~82 ft
  • Grapevine Pass (I-5): ~4,000 ft (steep climb)
  • Santa Ana: ~115 ft
  • Net elevation gain/loss is minimal, but the climb near Tejon Pass will impact consumption

Energy Use Estimate

Let’s assume:

• Efficiency: ~3.2 mi/kWh average (adjusted for elevation and HVAC)
• Usable range: ~282 miles from 100% to 10% SOC (88 kWh × 90% × 3.2 mi/kWh)

You’ll need at least one charging stop, possibly two depending on route and buffer.


Charging Strategy

Here’s a recommended route with optimized stops:

Segment	Distance	SOC Used	Stop Location	Charger Type	Charge To
San Jose → Kettleman City	~190 mi	~67%	Tesla Supercharger (NACS)	V3 (250 kW)	80%
Kettleman City → Santa Clarita	~130 mi	~45%	Tesla Supercharger	V3	80%
Santa Clarita → Santa Ana	~50 mi	~17%	No stop needed	—	Arrive ~13%

Notes:

• Kettleman City is a flagship Tesla site with amenities.
• Santa Clarita gives you buffer before LA traffic and elevation dips.
• You’ll arrive with ~13% SOC, staying above your 10% minimum.

Contingency & Buffer

• If HVAC usage spikes or traffic slows you down, consider Bakersfield or Tejon Ranch as alternate stops.
• You can simulate alternate SOC caps or swap Santa Clarita for Burbank if you prefer urban amenities.

Final Thoughts

This route keeps you well within your SOC constraints, leverages high-speed Tesla chargers via NACS, and accounts for elevation and weather. Want to simulate the return trip or explore alternate routes like US-101 coastal drive for scenery? I can also factor in hotel chargers or destination charging if you're staying overnight.


Let me know if you want a visual map or to tweak the SOC caps for faster travel.

 

[E90alex]

ABRP does it with just one stop. And a lot less typing.

 

[ChrisO]

I know, and it is about "caution", and those buffers it was talking about.

I update this with this prompt:
"Can this be done with one charge stop?"


Yes, Chris—it’s technically possible to make the trip from San Jose to Santa Ana with just one fast charge stop, but it’s a tight squeeze and depends on a few real-world variables. Let’s walk through the math and risk profile.


Range & Consumption Breakdown

• Battery: 88 kWh usable
• Efficiency estimate: ~3.2 mi/kWh (adjusted for elevation and HVAC)
• Usable range from 100% to 10% SOC:
  $$88 \times 0.9 \times 3.2 \approx 253 \text{ miles}$$
• Total trip distance: ~370 miles
  → You’ll need to cover ~117 miles after your first charge

One-Stop Strategy

Segment	Distance	SOC Used	Stop Location	Charge To
San Jose → Kettleman City	~190 mi	~67%	Tesla Supercharger (NACS)	80%
Kettleman City → Santa Ana	~180 mi	~63%	—	Arrive ~17%

Key Assumptions:

• You start at 100% SOC
• You charge to 80% at Kettleman City (~63 usable kWh)
• You maintain ~3.2 mi/kWh efficiency despite elevation climb at Grapevine

Risk Factors

• Elevation: Grapevine Pass (~4,000 ft) adds load
• Traffic: LA congestion can reduce efficiency
• HVAC: Mild temps tomorrow (~72–85°F) help, but AC use still matters
• Speed: Driving at limit is fine, but any aggressive acceleration will eat range

Buffer Analysis

• From Kettleman City to Santa Ana:
  • 180 mi ÷ 3.2 mi/kWh ≈ 56.25 kWh needed
  • You’ll have ~63 kWh available (80% SOC)
  • Buffer: ~6.75 kWh (~21 mi)

That’s just enough to stay above your 10% arrival threshold.


Verdict

Yes, you can do it with one stop at Kettleman City, but it’s a low-buffer plan. If you want more breathing room, a quick top-up at Santa Clarita or Tejon Ranch adds peace of mind without much time lost.


Want to simulate the return leg or test a 70% charge cap instead? I can run that next.

 

[ChrisO]

Note with other prompts I have seen it pick Bakerfield instead.

 

[ChrisO]

The main thing that I don't like about ABRP is that if I don't like their route/charge stations, I found it quite difficult to get it to pick a different route/charge stations.

With CoPilot I can just express that in a statement and it will update the plan.

 

[E90alex]

Note that it lists the same distance traveled and SOC used for both Kettleman City and Bakersfield.

I’ve driven between NorCal and SoCal countless times and I know for a fact those cities are not the same distance from San Jose.

So how much do you “trust” that the AI is not hallucinating and will choose what it says compared to a tool specifically designed for EV route planning and already accounts for various variables like vehicle efficiency and elevation changes?

 

[ChrisO]

Where do you see that it used the same SOC for Kettleman City and Bakersfield?

In both cases it used Kettleman as the first (or only charge point) with mentioning that Bakersfield as an alternate if one thought you can't make it.

Here is with the prompt for using Bakerfield instead and throw in the PlugShare ratings.

And yes, I would certainly be checking this on Google maps to make sure that it is a reasonable plan:



Perfect pivot, Chris. Let’s reroute your Mach-E trip using the Bakersfield Tesla Supercharger instead of Kettleman City, while keeping your SOC constraints, weather, and elevation in play.


Updated Trip Plan: San Jose → Santa Ana via Bakersfield

Segment	Distance	SOC Used	Stop Location	Charge To	Notes
San Jose → Bakersfield	~240 miles	~85%	Bakersfield Tesla Supercharger	80%	PlugShare rating:
Bakersfield → Santa Ana	~130 miles	~45%	—	—	Arrive with ~35% SOC

Efficiency & Range Recap

• Battery: 88 kWh usable
• Efficiency: ~3.2 mi/kWh (adjusted for mild weather and elevation)
• Usable range (100% → 10%): ~253 miles
• Bakersfield arrival SOC: ~15%–20% (safe buffer)
• Charge to 80%: ~250 miles of usable range
• Santa Ana arrival SOC: ~35% (well above 10% minimum)

Weather & Elevation Impact

• Tomorrow’s Weather:
  • San Jose: ~72°F, sunny
  • Bakersfield: ~85°F, dry heat
  • Santa Ana: ~78°F, mild breeze
• Elevation:
  • Climb through Pacheco Pass (~1,368 ft)
  • Descend into Central Valley
  • Grapevine Pass (~4,000 ft) adds load, but regen on descent helps
  • Net elevation gain is minimal

Charging Stop Details

Bakersfield Tesla Supercharger

• Location: South Enos Lane
• Power: Up to 150 kW
• PlugShare Rating:
• Amenities: Food, restrooms, nearby Chevron & Starbucks

Risk Buffer & Contingency

• You’ll arrive in Santa Ana with ~35% SOC, giving you ~75 miles of buffer
• If HVAC load or traffic increases consumption, fallback options include:
  • Tejon Ranch Supercharger (~40 miles before Bakersfield)
  • Santa Clarita Supercharger (~60 miles before Santa Ana)

This route gives you a solid buffer, avoids overcharging, and uses a top-rated NACS-compatible site. Want to simulate the return leg or test a 70% charge cap for faster stops? I can run that next.

 

[E90alex]

Sorry I read that wrong. I thought the one stop plan was stopping in Bakersfield b it it still has you stopping in Kettleman City.

Doesn’t make sense to make one stop in Kettleman City with an estimated arrival SOC of ~33%. Not optimal at all.

Bakersfield would be the best stop with 17% according to ABRP which is still plenty buffer. You could do Lost Hills if you wanted to be more conservative or try to push to Tejon Ranch if you wanted to push it.

 

[ChrisO]

I tried it with using the Telsa rating and it noted that even though that location picked has 5 stars there are "cow smells", so I said to avoid them, and it switched to Tejon Ranch.

That is what I like about doing it this way, trying different things is so much more natural than trying to get ABRP to do it.

 

[ChrisO]

At the end of the day, I would certainly be on the conservative side for my actual trip. But it is interesting to play with to see different possibilities in a natural, quick manner.

 

[leehinde]

I just asked ChatGPT about a trip to Santa Cruz and it recommended the Tesla Supercharger at Gilroy, which isn't open to non-teslas. Trust (or not) but verify.

 

[ChrisO]

I just asked ChatGPT about a trip to Santa Cruz and it recommended the Tesla Supercharger at Gilroy, which isn't open to non-teslas. Trust (or not) but verify.

And another tip I would have is if a gets something wrong point it out. That should help it “learn”. And you can add it into the prompt to check that.

one has to think about it like talking to someone that might not have considered all the needed details. But in this case saying that you have a NACS adapter should have been enough.

 

[RickMachE]

AI is so wonderful....

Sponsored