Driving Gen

This environment is part of the Grid World environments. Please read that page first for general information.


Possible Agents	(‘0’, ‘1’)
Action Spaces	{‘0’: Discrete(5), ‘1’: Discrete(5)}
Observation Spaces	{‘0’: Tuple(Tuple(Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4)), Discrete(4), Tuple(Discrete(14), Discrete(14)), Tuple(Discrete(14), Discrete(14)), Discrete(2), Discrete(2)), ‘1’: Tuple(Tuple(Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4), Discrete(4)), Discrete(4), Tuple(Discrete(14), Discrete(14)), Tuple(Discrete(14), Discrete(14)), Discrete(2), Discrete(2))}
Symmetric	True
Import	`posggym.make("DrivingGen-v1")`

The Generated Driving Grid World Environment.

This is the same as the Driving Environment except that a new grid is generated at each reset.

For environment attributes see Driving environment class documentation.

Arguments

num_agents - the number of agents in the environment (default = 2).
obs_dim - the local observation dimensions, specifying how many cells in front, behind, and to each side the agent observes (default = (3, 1, 1), resulting in the agent observing a 5x3 area: 3 in front, 1 behind, 1 to each side.)
generator_params - the parameters to use for generating the grid (default = “14x14”). This can either be a string specyfing one of the supported sets of params (can be “7x7”, “14x14”, “28x28”) or a dictionary with the following keys:
- width - width of the grid
- height - height of the grid
- max_obstacle_size - max size of an obstacle
- max_num_obstacles - maximum number of obstacles in the grid
n_grids - the number of different grids to generate, if provided then n_grids will be generated and these will be cycled through for each reset. If None then a different, possibly unique, grid will be generated each episode (default = None).
shuffle_grid_order - whether to shuffle the order in which different grids appear. Only has any effect if n_grids > 1 (default = True).

The DrivingGen environment comes with a number of pre-built sets of grid generator parameters which can be passed as an argument to posggym.make:

Name	Grid size	max obstacle size	max num obstacles
`7x7`	7x7	2	21
`14x14`	14x14	3	42
`28x28`	28x28	4	84

For example to use the DrivingGen environment with the pre-build 7x7 set of generation parameters and 2 agents, you would use:

import posggym
env = posggym.make('DrivingGen-v1', generator_params="7x7", num_agents="2")

v1: Updated to Driving-v1 (See Version history of Driving environment for details).
v0: Initial version