...
And let's take a look at the 2nd blue group of agents. We place them like the below picture and make it follow the terrain.
And we also, we need to build some scene perception elements for make the agents can feel the environment correctly.
We want our agent . Their run along the terrain, and then climb to the hill toward the target spot, then if the agent reach the top and enter the bound, it will turn to dynamics ragdoll and push by a small force outside the hill.
And their logic setup is:
- On the ground (zone 1)
- run toward the spot
- follow the terrain
- avoid each other
- On the hill (zone 2)
- climb toward the top of hill
- climb with different actions
- when enter into the bound, enable dynamics
- there is a push force field, push the dyanmics agent outward a little bit.
And we can simulate and cache 700 frames, and start shooting it from the frame 500:
