Merge

In this task, the ego-vehicle starts on a main highway but soon approaches a road junction with incoming vehicles on the access ramp. The agent’s objective is now to maintain a high speed while making room for the vehicles so that they can safely merge in the traffic.

Usage

env = gym.make("merge-v0")

Default configuration

{
    "observation": {
        "type": "TimeToCollision"
    },
    "action": {
        "type": "DiscreteMetaAction"
    },
    "simulation_frequency": 15,  # [Hz]
    "policy_frequency": 1,  # [Hz]
    "other_vehicles_type": "highway_env.vehicle.behavior.IDMVehicle",
    "screen_width": 600,  # [px]
    "screen_height": 150,  # [px]
    "centering_position": [0.3, 0.5],
    "scaling": 5.5,
    "show_trajectories": False,
    "render_agent": True,
    "offscreen_rendering": False
}

More specifically, it is defined in:

classmethod MergeEnv.default_config() → dict

Default environment configuration.

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

API

class highway_env.envs.merge_env.MergeEnv(config: dict = None)

A highway merge negotiation environment.

The ego-vehicle is driving on a highway and approached a merge, with some vehicles incoming on the access ramp. It is rewarded for maintaining a high speed and avoiding collisions, but also making room for merging vehicles.

COLLISION_REWARD: float = -1

RIGHT_LANE_REWARD: float = 0.1

HIGH_SPEED_REWARD: float = 0.2

MERGING_SPEED_REWARD: float = -0.5

LANE_CHANGE_REWARD: float = -0.05

_reward(action: int) → float

The vehicle is rewarded for driving with high speed on lanes to the right and avoiding collisions

But an additional altruistic penalty is also suffered if any vehicle on the merging lane has a low speed.

Parameters

action – the action performed

Returns

the reward of the state-action transition

_is_terminal() → bool

The episode is over when a collision occurs or when the access ramp has been passed.

reset() → numpy.ndarray

Reset the environment to it’s initial configuration

Returns

the observation of the reset state

_make_road() → None

Make a road composed of a straight highway and a merging lane.

Returns

the road

_make_vehicles() → None

Populate a road with several vehicles on the highway and on the merging lane, as well as an ego-vehicle.

Returns

the ego-vehicle

__annotations__ = {'COLLISION_REWARD': <class 'float'>, 'HIGH_SPEED_REWARD': <class 'float'>, 'LANE_CHANGE_REWARD': <class 'float'>, 'MERGING_SPEED_REWARD': <class 'float'>, 'RIGHT_LANE_REWARD': <class 'float'>}

__module__ = 'highway_env.envs.merge_env'