public Dictionary <State, float> GetPossibleStatesFromIntent(State currentState, GridWorldIntent gridWorldIntent)
{
Dictionary <State, float> possibleStates = new Dictionary <State, float>();
possibleStates.Add(GetNextState(currentState, gridWorldIntent), 1.0f);
return(possibleStates);
}
public bool PolicyImprovement()
{
bool policyStable = true;
foreach (var currentState in _allStates)
{
if (GameSelector.type == GameSelector.GameType.GridWorld)
{
GridWorldIntent tempPolicy = currentState.gridWorldPolicy;
currentState.gridWorldPolicy = GetBestIntent(currentState);
if (tempPolicy != currentState.gridWorldPolicy)
{
policyStable = false;
}
}
else if (GameSelector.type == GameSelector.GameType.Sokoban)
{
SokobanIntent tempPolicy = currentState.sokobanPolicy;
currentState.sokobanPolicy = GetBestIntentSokoban(currentState);
if (tempPolicy != currentState.sokobanPolicy)
{
policyStable = false;
}
}
}
if (!policyStable)
{
PolicyEvaluation();
}
return(policyStable);
}
public void InitializePolicyIteration()
{
Debug.Log("Initialization");
_allStates = new List <State>();
if (GameSelector.type == GameSelector.GameType.GridWorld)
{
for (int i = 0; i < gridWorldController.grid.gridHeight; ++i)
{
for (int j = 0; j < gridWorldController.grid.gridWidth; ++j)
{
State currentState = new State();
currentState.currentPlayerPos = new Vector3(i, 0, j);
if (currentState.currentPlayerPos == gridWorldController.grid.endPos)
{
currentState.stateValue = 1000.0f;
}
else
{
currentState.stateValue = 0;
}
_allStates.Add(currentState);
}
}
Debug.Log("Number of states : " + _allStates.Count);
foreach (var currentState in _allStates)
{
GridWorldIntent wantedGridWorldIntent = (GridWorldIntent)Random.Range(0, 3);
currentState.gridWorldPolicy = wantedGridWorldIntent;
}
}
}
public bool CheckIntent(State currentState, GridWorldIntent wantedGridWorldIntent)
{
if (GetNextState(currentState, wantedGridWorldIntent) == null)
{
return(false);
}
if (GetCellType(GetNextState(currentState, wantedGridWorldIntent).currentPlayerPos) == CellType.Obstacle)
{
return(false);
}
return(true);
}
public GridWorldIntent GetBestIntent(State currentState)
{
float max = float.MinValue;
GridWorldIntent bestGridWorldIntent = currentState.gridWorldPolicy;
for (int i = 0; i < 4; ++i)
{
if (CheckIntent(currentState, (GridWorldIntent)i))
{
State tempState = GetNextState(currentState, (GridWorldIntent)i);
if (tempState.stateValue > max)
{
max = tempState.stateValue;
bestGridWorldIntent = (GridWorldIntent)i;
}
}
}
return(bestGridWorldIntent);
}
public State GetNextState(State currentState, GridWorldIntent gridWorldIntent)
{
Vector3 nextPlayerPos = Vector3.zero;
switch (gridWorldIntent)
{
case GridWorldIntent.Down:
nextPlayerPos = currentState.currentPlayerPos - Vector3.forward;
break;
case GridWorldIntent.Up:
nextPlayerPos = currentState.currentPlayerPos + Vector3.forward;
break;
case GridWorldIntent.Left:
nextPlayerPos = currentState.currentPlayerPos + Vector3.left;
break;
case GridWorldIntent.Right:
nextPlayerPos = currentState.currentPlayerPos - Vector3.left;
break;
}
return(GetStateFromPos(nextPlayerPos));
}