public override void Learn(GridWorld world)
{
float[] currentState = world.GetState();
float currentReward = CurrentReward();
learningAgent.Learn(currentState, currentReward);
}
public override WorldAction Process(GridWorld world)
{
Reward(-1);
float[] state = world.GetState();
int index = learningAgent.Process(state);
return(IndexToAction(index));
}
public override WorldAction Process(GridWorld world)
{
foreach (WorldObject obj in world.objects)
{
if (obj == this)
{
continue;
}
if ((obj.pos.X == this.pos.X) && (obj.pos.Y == this.pos.Y))
{
obj.Reward(reachReward);
return(WorldAction.END);
}
//obj.Reward(-1);
}
return(WorldAction.None);
}
public virtual void Learn(GridWorld world)
{
}
public virtual WorldAction Process(GridWorld world)
{
return(0);
}