Continuous Mountain Car

lerax.env.ContinuousMountainCar

Bases: AbstractClassicControlEnv[ContinuousMountainCarState, Float[Array, ''], Float[Array, '2']]

Continuous Mountain Car environment matching the Gymnasium Continuous MountainCar environment.

Note

To achieve identical dynamics to Gymnasium set solver=diffrax.Euler().

The action space is a 1-dimensional continuous space representing the force applied to the car in the range [-1.0, 1.0].

The observation space is a 2-dimensional continuous space representing the position and velocity of the car:

Index	Observation	Min Value	Max Value
0	Car Position	-1.2	0.6
1	Car Velocity	-0.07	0.07

Parameters:

Name	Type	Description	Default
`min_action`	`Float[ArrayLike, '']`	Minimum action value (default: -1.0).	`-1.0`
`max_action`	`Float[ArrayLike, '']`	Maximum action value (default: 1.0).	`1.0`
`min_position`	`Float[ArrayLike, '']`	Minimum position of the car (default: -1.2).	`-1.2`
`max_position`	`Float[ArrayLike, '']`	Maximum position of the car (default: 0.6).	`0.6`
`max_speed`	`Float[ArrayLike, '']`	Maximum speed of the car (default: 0.07).	`0.07`
`goal_position`	`Float[ArrayLike, '']`	Position at which the goal is reached (default: 0.5).	`0.5`
`power`	`Float[ArrayLike, '']`	Power of the car's engine (default: 0.0015).	`0.0015`
`dt`	`Float[ArrayLike, '']`	Time step for each action (default: 1.0).	`1.0`
`solver`	`diffrax.AbstractSolver \| None`	Diffrax solver to use for ODE integration (default: Tsit5).	`None`
`stepsize_controller`	`diffrax.AbstractStepSizeController \| None`	Step size controller for adaptive solvers (default: PIDController with rtol=1e-5, atol=1e-5).	`None`