Crops the image, then resize it back to the original shape, finally applies horizontal flipping.

Define which sensors should be equipped to the vehicle

Tells whether or not the agent has reached the goal (destination) location

Removes the previously spawned actors (vehicles and pedestrians/walkers)

Draw a list of waypoints at a certain height given in z. :param world: carla.worlds object :param waypoints: list or iterable container

Defined the shared-network architecture, returns its last layer

Implicit Dynamics Model, - Inputs: state/observation, action, value

Used during learning `learn(...)` to terminate an episode

Ends a sequence of transitions {(s, a, r, d, i)_t}

Gaussian blur augmentation as defined in SimCLR paper

Computes batches of data for updating the policy network

Computes batches of data for updating the value network

Translates a 3D point from global to local coordinates using the current transformation as reference

Check if a target object is within a certain distance from a reference object. A vehicle in front would be something around 0 deg, while one

Callback. Triggers when a world's 'tick' occurs, meaning that data from sensors are been collected because a simulation step of the CARLA's wo

This module is in charge of warning in case of a collision with any pedestrian. :param location: current location of the

Calculates the sequence of waypoints (a route) from origin to destination.

Colormaps: https://matplotlib.org/tutorials/colors/colormaps.html

Defines which data to use in `get_policy_batches()`

PPO-Clip Objective

Return the given number if positive, else 0 :param num: value to check

Resize the given image. :param image: a numpy array with shape (height, width, channels). :param size: (width, height) to resize the i

Reward function

Agent's reward function

Execute one step of navigation. :param debug: boolean flag for debugging :return: control

Writes to file (npz - numpy compressed format) all the transitions collected so far

Sets new global plan. :param current_plan: list of waypoints in the actual plan

Request new target speed. :param speed: new target speed in km/h

Performs initial server data lookup for detailed topology and builds graph representation of the worlds map.

Performs initial server data lookup for detailed topology and builds graph representation of the world map.

Stage-5: town with dense traffic (100 vehicles and 200 pedestrians) + random light weather + data-aug

Collects one transition (s, a, r, d, i)

Networks summary

Tells whether the episode is terminated or not.

Normalizes some tensor x to 0-mean 1-stddev

Maps a discrete array of bins into their corresponding continuous values

This method is in charge of behaviors for red lights and stops. WARNING: What follows is a proxy to avoid having a car brake after r

Returns the unit vector from location1 to location2.

Unregisters a given [callback] to a specific [event]

Stores the given variables in the configuration dict for later saving

This method updates the information regarding the ego vehicle based on the surrounding world.