Hi All. First time poster here.
I have a dilemma. I am putting together a simple game which include drivable vehicles which also drive themselves. I have a parking area which I want the cars to enter, pick a space and park in, and also reverse out of the space and leave the parking area.
So far as I know I have two main options for implementation. ML-agents or Waypoints. I’ve only just started looking into ML-Agents but it feels a little heavy handed and possibly a little gimmicky for what I need it to do, but on the other hand, creating a waypoint path for every parking space also seems like a lot of grind work. Truth be told, whilst I consider myself a good coder, I don’t know enough about ML-Agents to make an informed decision about which would be the best in the long run.
So I’m reaching out to people who might have a better idea of the nitty gritty who can perhaps offer some convincing arguments for either case (or offer a better alternative).
The vehicles need to be able to avoid obstacles such as other vehicles (moving or parked)
I know that I could set up a system by which each individual space has a “route” consisting of a list of waypoints which the vehicle would follow from the entrance to park and then another route to exit. Which is currently what I’m considering doing.
Any thoughts?