Map the pygame keyboard events to control decisions :param action_type: the ActionType that defines how the vehicle is controlled

Check if rect1 has a corner inside rect2 :param rect1: (center, length, width, angle) :param rect2: (center, length, width, angle)

Compute the interval of an integrator system: dx = -k*x :param x: state interval :param k: gain interval, must be positive :return:

Compute the negative part of an interval :param a: interval [a_min, a_max] :return: the interval of its negative part min(a, 0)

Scale an intervals :param a: matrix a :param b: interval [b_min, b_max] :return: the interval of their product ab

State: [lateral speed v, yaw rate r] :return: lateral dynamics A, B

State: [lateral speed v, yaw rate r] :return: lateral dynamics A0, phi, B such that dx = (A0 + theta^T phi)x + B u

Step the interval observer dynamics :param dt: timestep [s]

Predict the future positions of the vehicle along its planned route, under constant speed :param times: timesteps of prediction

Control the speed of the vehicle. Using a simple proportional controller. :param target_speed: the desired speed :re

Perform an action and step the environment dynamics. The action is executed by the ego-vehicle, and all other vehicles on the road p

The class of a vehicle able to execute the action. Must return a subclass of :py:class:`highway_env.vehicle.kinematics.Vehicle`.

New sinusoidal lane. :param start: the lane starting position [m] :param end: the lane ending position [m] :param am

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

Automatically render the intermediate frames while an action is still ongoing. This allows to render the whole video and not only si

A constraint metric, for budgeted MDP. If a constraint is defined, it must be used with an alternate reward that doesn't contain it

Make a road composed of a straight highway and a merging lane. :return: the road

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

Reset the scene: roads and vehicles. This method must be overloaded by the environments.

Return the reward associated with performing a given action and ending up in the current state. :param action: the last action perfo

Perform several steps of simulation with constant action.

Store an action to be repeated. :param action: the input action

Execute an action. For now, no action is supported because the vehicle takes all decisions of acceleration and lane changes

Breadth-first search of all routes from start to goal. :param start: starting node :param goal: goal node :return: l

Many thanks to https://stackoverflow.com/a/54714144.

Decide when to change lane. Based on: - frequency; - closeness of the target lane; - MOBIL model.

Decide when to change lane. Based on: - frequency; - closeness of the target lane; - MOBIL model.

para: vehicle: the ego vehicle surrounding_vehicles: [v_fl, v_rl, v_fr, v_rr] env_copy: copy of self

Store features and outputs for parameter regression.

Compute a confidence ellipsoid over the parameter theta, where y = theta^T phi :param data: a dictionary {"features": [phi_0,...,phi_N], "ou

Draw a striped line on one side of a lane, on a surface. :param lane: the lane :param surface: the pygame surface :p

Draw a continuous line on one side of a lane, on a surface. :param lane: the lane :param surface: the pygame surface

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

Display the road and vehicles on a pygame window.

Display the whole trajectory of a vehicle on a pygame surface. :param vehicle: the vehicle states within the trajectory to be displa

Display the road objects on a surface. :param road: the road to be displayed :param surface: the pygame surface :par

Display the road vehicles on a surface. :param road: the road to be displayed :param surface: the pygame surface :pa

Display the whole trajectory of a vehicle on a pygame surface. :param states: the list of vehicle states within the trajectory to be

Compute a weighted distance in position and heading to the lane.

Find conflicts and resolve them by assigning yielding vehicles and stopping them.

Time-To-Collision (TTC) representation of the state. The state reward is defined from a occupancy grid over different TTCs and lanes. The gr

At the end of a lane, automatically switch to a next one.

At the end of a lane, automatically switch to a next one.

State: [position y, yaw psi, lateral speed v, yaw rate r] The system is linearized around psi = 0 :return: lateral dynamics

State: [position y, yaw psi, lateral speed v, yaw rate r] The system is linearized around psi = 0 :return: lateral dynamics

Get the list of lanes that could be followed by this vehicle. :param lane_change_model: - model: assume that the vehicle will follow

the rendered image as a rbg array.

Handle pygame events by forwarding them to the display and environment vehicle.

Get the lane heading at a given longitudinal lane coordinate. :param longitudinal: longitudinal lane coordinate [m] :return:

Converts an interval in local (longiturinal, lateral) coordinates to an interval in absolute x,y coordinates :param longitudinal_i: the long

Is the lane 2 leading to a road within lane 1's route? Vehicles on these lanes must be considered for collisions. :param lan

Check whether a dataset {phi_n, y_n} is consistent The last observation should be in the confidence ellipsoid obtained by the N-1 first obse

Is lane 1 leading to of lane 2?

Is lane 1 in the same road as lane 2?