U-Turn¶

A U-turn risk analysis task: the agent overtakes vehicles that are blocking the traffic while navigating a double-lane U-turn. Six strategically placed vehicles force the agent to balance high-speed overtaking with ensuring safety.

Usage¶

env = gym.make("u-turn-v0")

Versions¶

ID	Description
`u-turn-v0`	Initial version. Same-segment neighbour search only. Preserved for reproducibility.
`u-turn-v1`	Connected-lane neighbour search enabled by default. Recommended for new experiments.

See Neighbour vehicles for details.

Default configuration¶

{
    "observation": {
        "type": "TimeToCollision",
        "horizon": 16,
    },
    "action": {
        "type": "DiscreteMetaAction",
        "target_speeds": [8, 16, 24],
    },
    "screen_width": 789,
    "screen_height": 289,
    "duration": 10,  # [s]
    "collision_reward": -1.0,
    "left_lane_reward": 0.1,
    "high_speed_reward": 0.4,
    "reward_speed_range": [8, 24],
    "normalize_reward": True,
    "offroad_terminal": False,
}

More specifically, it is defined in:

classmethod UTurnEnv.default_config() → dict[source]

Default environment configuration.

Can be overloaded in environment implementations, or by calling configure(). :return: a configuration dict

API¶

class highway_env.envs.u_turn_env.UTurnEnv(config: dict = None, render_mode: str | None = None)[source]¶

U-Turn risk analysis task: the agent overtakes vehicles that are blocking the traffic. High speed overtaking must be balanced with ensuring safety.

classmethod default_config() → dict[source]¶

Default environment configuration.

Can be overloaded in environment implementations, or by calling configure(). :return: a configuration dict