From what I can glean from the article, it may not be possible to completely address the problem in software, because weather prediction is inexact, and it's hard to account for all eventualities. The software can say, "100 miles of range unless the temperature drops more than 5 degrees from forecast, or unless it turns out your hotel has no indoor parking, or if you have to stay late at work and your car spends an extra four hours outside, or your flight is cancelled and your car spends an extra day at the airport." That's not to say it can't be improved, though.
That is sort of what I was wondering but recognize such systems are tricky given the feedback. Also turning electricity into heat is hugely inefficient on cold drives, if I were Tesla I would consider a 'cold weather' kith where you could put a bottle of propane in the trunk or something which could run a catalytic heater without consuming any battery power except for recirculating fans.
I was thinking of software only solutions, but I think a hardware fix may be best.
I wonder if Tesla chose to optimize for other scenarios in the design of the car: Tesla distributes the battery across the entire body, which makes for great handling, fault tolerance, and safety, but may make it more difficult to insulate than in other cars, which typically just lump the battery in the back, surrounded by insulation.
An "inside" battery also takes advantage of the solar greenhouse effect from the glass-enclosed interior. On cold-but-sunny day, a parked car will heat up substantially simply due to the ~1kw/m^2 of solar energy delivered to the vehicle. Of course, you don't want the battery getting to hot, either. In addition to better thermal insulation, perhaps Tesla could add solar panels to the roof to aid in heating and cooling.
I hope this problem is solved, as it would be a bummer not to be able to take a Tesla on a ski trip.
