public Form1()
{
InitializeComponent();
state = new double[] { -Math.PI, 0 };
sim = new InvertedPendulumEnvironmentSimulator();
timer1.Start();
}
public RL_TDlambda()
{
InitializeComponent();
sim = new InvertedPendulumEnvironmentSimulator();
igmn1 = new IGMN(new Vector(new double[] { 1, 1, 1 }));
igmn2 = new IGMN(new Vector(new double[] { 1, 1, 1 }));
tab1 = new TabularApproximator(new int[] { 200, 200 }, new double[] { -Math.PI, -1.5*Math.PI }, new double[] { Math.PI, 1.5 * Math.PI });
tab2 = new TabularApproximator(new int[] { 200, 200 }, new double[] { -Math.PI, -1.5*Math.PI }, new double[] { Math.PI, 1.5 * Math.PI });
r = new Random();
isRunning = true;
trainingThread = new Thread(new ThreadStart(train));
trainingThread.Start();
timer1.Start();
}
protected Bitmap VisualizeInvertedPendulum()
{
int res = 1200;
Bitmap bm = new Bitmap(res, res);
Graphics g = Graphics.FromImage(bm);
InvertedPendulumEnvironmentSimulator simm = new InvertedPendulumEnvironmentSimulator();
double[] state = new double[] { Math.PI, 0 };
double sumreward = 0;
for (int i = 0; i < 300; ++i)
{
double M;
double q_act;
double reward;
GetAction(state, out M, out q_act, true);
simm.Simulate(state, new double[] { M }, dT, out state, out reward);
sumreward += reward;
if (i % 1 == 0)
{
float zx = 75 + ((i % 100)) * 10;
float zy = 150 + 150 * (i / 100);
float z = 50;
float e1x = zx + (float)(z * simm.Length * Math.Sin(simm.Angle));
float e1y = zy - (float)(z * simm.Length * Math.Cos(simm.Angle));
Pen p = new Pen(Color.Black);
g.DrawLine(p, zx, zy, e1x, e1y);
g.DrawEllipse(p, zx-2, zy-2, 4, 4);
}
}
if (bestsumreward < sumreward)
{
bestsumreward = sumreward;
bm.Save("invertedpendulum.png");
}
Console.Out.WriteLine(sumreward);
return bm;
}