It works as you can see there people who have used such features. There can be some issues that can limit how well it works.
In a normal packet node, you would receive the signal and feed the audio into a TNC and decode the data. That would get sent to another TNC which would convert it back to tones to send to the transmitter. The results is that some marginal signals (weak, noisy, distorted etc.) can still produce audio that decodes. If you simply repeat that then you may introduce more distortions resulting in a signal that can never be decoded. 2x TNC solution is better but more expensive. You can also put some rules in so only your packets are repeated rather than anything heard.
Secondly, the radio doing the repeating needs to be able to unsquelch itsekf quick and get repeating the incoming signal. If it is a bit slow, the initial data received may not get repeated. You can fix that by making the portable signal have a longer TX delay if needed.
However, as Carolyn says, if the car is parked at the bottom of the hill then it’s not too likely to be able to get a signal into the APRS network. So whilst you can probably get a simple audio repeater to operate you may not gain anything. If you go to HF on one side then you really will need a pair of back to back TNCs as HF is done at slower speeds than VHF. 300bps AFSK is one standard but you may find lots of HF APRS stations adopting more modern modulation schemes because the work better. At this point as long as you can be heard by your repeater then you can get your local VHF/UHF APRS packets out via HF.
The question you need to ask is whether the setting up and experimenting with these ideas is the fun and worthwhile or are you trying to get a message through saying “I am at the summit on freq X” i.e. a self spot. If it’s the latter, simply use a smartphone with APRSdroid or some such to send the APRS message and connect that to your HF rig at the summit and send your message then.
If the message being received is more important that the experimenting with a portable repeater etc. then there are other ways much more likely to get the message through. But they probably wont use amateur radio but commercial services that cost money. You need to decide if the message is the purpose (in which case spend money to send it) or the “doing of the experiment” is the purpose and then playing with radios and TNCs etc.
If you have a radio that does x-band repeating already you can try the simple method for not much effort. When I though about it I came to the conclusion that I want to get the spot message through rather than play with APRS nodes so I spent money on a commercial service. I still had “fun” (no I didn’t) writing the software to make it work. But now it boils down to a spot costing 40p to send. The fun comes from walking in the countryside and being the DX when I set up the SOTA station.