mlpack::rl::Pendulum Class Reference

Implementation of Pendulum task. More...

#include <pendulum.hpp>

Classes
class	Action
	Implementation of action of Pendulum. More...
class	State
	Implementation of state of Pendulum. More...

Public Member Functions
	Pendulum (const size_t maxSteps=200, const double maxAngularVelocity=8, const double maxTorque=2.0, const double dt=0.05, const double doneReward=0.0)
	Construct a Pendulum instance using the given values. More...
double	Sample (const State &state, const Action &action, State &nextState)
	Dynamics of Pendulum. More...
double	Sample (const State &state, const Action &action)
	Dynamics of Pendulum. More...
State	InitialSample ()
	Initial theta is randomly generated within [-pi, pi]. More...
double	AngleNormalize (double theta) const
	This function calculates the normalized angle for a particular theta. More...
bool	IsTerminal (const State &) const
	This function checks if the pendulum has reaches a terminal state. More...
size_t	StepsPerformed () const
	Get the number of steps performed.
size_t	MaxSteps () const
	Get the maximum number of steps allowed.
size_t &	MaxSteps ()
	Set the maximum number of steps allowed.

Detailed Description

Implementation of Pendulum task.

The inverted pendulum swingup problem is a classic problem in the control literature. In this version of the problem, the pendulum starts in a random position, and the goal is to swing it up so it stays upright

Constructor & Destructor Documentation

Pendulum()

mlpack::rl::Pendulum::Pendulum ( const size_t  maxSteps = 200, const double  maxAngularVelocity = 8, const double  maxTorque = 2.0, const double  dt = 0.05, const double  doneReward = 0.0  )

inline

Construct a Pendulum instance using the given values.

Parameters

maxSteps	The number of steps after which the episode terminates. If the value is 0, there is no limit (Default: 200 steps).
maxAngularVelocity	Maximum angular velocity.
maxTorque	Maximum torque.
dt	The differential value.
doneReward	The reward recieved by the agent on success.

Member Function Documentation

AngleNormalize()

double mlpack::rl::Pendulum::AngleNormalize ( double  theta ) const

inline

This function calculates the normalized angle for a particular theta.

Parameters

theta

The un-normalized angle.

InitialSample()

State mlpack::rl::Pendulum::InitialSample ( )

inline

Initial theta is randomly generated within [-pi, pi].

Initial angular velocity is randomly generated within [-1, 1].

Returns

Initial state for each episode.

IsTerminal()

bool mlpack::rl::Pendulum::IsTerminal ( const State &  ) const

inline

This function checks if the pendulum has reaches a terminal state.

Parameters

*	(state) desired state.

Returns

true if state is a terminal state, otherwise false.

Sample() [1/2]

double mlpack::rl::Pendulum::Sample ( const State &  state, const Action &  action, State &  nextState  )

inline

Dynamics of Pendulum.

Get reward and next state based on current state and current action.

Parameters

state	The current state.
action	The current action.
nextState	The next state.

Returns

reward, The reward for taking the action taken for current state.

Sample() [2/2]

double mlpack::rl::Pendulum::Sample ( const State &  state, const Action &  action  )

inline

Dynamics of Pendulum.

Get reward based on current state and current action

Parameters

state	The current state.
action	The current action.

Returns

reward, The reward.

---

The documentation for this class was generated from the following file:

• src/mlpack/methods/reinforcement_learning/environment/pendulum.hpp