Since the alternator is an AC source, we get pulsed current behind the rectifier. The battery can fill the gaps between the pulses. My first attempt would be a simple parallel ciruit with an additional diode to avoid a charge current flowing from the alternator back to the battery.
This may work perfectly and is worth an experiment.
V1 is the alternator, the simulation is not very realistic, because its inductance is ignored here. If I add an inductor in series with V1, the simulator dies.
simulate this circuit – Schematic created using CircuitLab
If the above is not sufficient, a true current limiter for the battery side is possible. This is not very efficient, the MOSFET M1 must dissipate 12 W worst case. The circuit limits the battery current to 1.2A during the gaps. During the peak current pulses from the alternator, the battery current drops to zero. So the average load on the battery is a bit lower than 15W. A lower value for R4 can allow more battery current, this can be adjusted in a real world circuit.
simulate this circuit
D2 is used as a cheap voltage reference and the voltage divider R4/R5 sets the target current for the battery.
A more efficient solution would involve a current limited SMPS stage, I will draw one, when I have some more time.