Driving

These policies are for the Driving environment. Read environment page for detailed information about the environment.

Generic

These policies can be used for any version of this environment.

env = posggym.make("Driving-v1")

Policy	ID	Valid Agent IDs	Description
A0Shortestpath	Driving-v1/A0Shortestpath-v0	All	Follows shortest path to destination, doesn’t go full speed, and stops if it observes another a car at any distance (aggressiveness=0.0)
A40Shortestpath	Driving-v1/A40Shortestpath-v0	All	Follows shortest path to destination, doesn’t go full speed, and stops if it observes a car near edge of observation range (aggressiveness=0.40)
A60Shortestpath	Driving-v1/A60Shortestpath-v0	All	Follows shortest path to destination, goes up to full speed, and stops if it observes a car a medium distance away (aggressiveness=0.60)
A80Shortestpath	Driving-v1/A80Shortestpath-v0	All	Follows shortest path to destination, goes up to full speed, and stops if it observes a car a very short distance away (aggressiveness=0.80)
A100Shortestpath	Driving-v1/A100Shortestpath-v0	All	Follows shortest path to destination, goes up to full speed, and ignores other vehicles (aggressiveness=1.0)

grid=14x14RoundAbout-num_agents=2

env = posggym.make(
    "Driving-v1",
    grid="14x14RoundAbout",
    num_agents=2,
    obs_dim=(3, 1, 1)
)

Policy	ID	Valid Agent IDs	Description
RL1	Driving-v1/grid=14x14RoundAbout-num_agents=2/RL1-v0	All	Deep RL policy trained using PPO and self-play.
RL2	Driving-v1/grid=14x14RoundAbout-num_agents=2/RL2-v0	All	Deep RL policy trained using PPO and self-play.
RL3	Driving-v1/grid=14x14RoundAbout-num_agents=2/RL3-v0	All	Deep RL policy trained using PPO and self-play.
RL4	Driving-v1/grid=14x14RoundAbout-num_agents=2/RL4-v0	All	Deep RL policy trained using PPO and self-play.
RL5	Driving-v1/grid=14x14RoundAbout-num_agents=2/RL5-v0	All	Deep RL policy trained using PPO and self-play.