Action Mask and working cozmo

This commit is contained in:
Tobi
2019-06-20 17:32:18 +02:00
parent 8bb5da8946
commit ecd9ad491d
21 changed files with 25226 additions and 370 deletions
+65 -6
View File
@@ -27,6 +27,9 @@ public class CozmoAgent : Agent
private ImageProcessor imageProcessor; // reference to the ImageProcessor
private int nearAreaLimit = 0; // X coordinate limit for the near to the imagecenter area
private int centerOfImageX = 0; // Middle of the image in x direction
private MovementState lastChosenMovement = MovementState.Stop; // The last action/movement that was executed
private double startTime = Time.time;
private void Start()
{
@@ -43,9 +46,46 @@ public class CozmoAgent : Agent
}
public override void CollectObservations()
{
SetMask();
}
// Set ActionMask for training
private void SetMask()
{
switch (lastChosenMovement)
{
// Do not allow stop decision after a stop
case (MovementState.Stop):
SetActionMask(STOP);
break;
// Do not allow stop after forward
case (MovementState.Forward):
SetActionMask(STOP);
break;
// Do not allow stop & left after right
case (MovementState.Right):
SetActionMask(STOP);
SetActionMask(LEFT);
break;
// Do not allow stop & right after left
case (MovementState.Left):
SetActionMask(STOP);
SetActionMask(RIGHT);
break;
default:
throw new ArgumentException("Invalid MovementState.");
}
}
// to be implemented by the developer
public override void AgentAction(float[] vectorAction, string textAction)
{
double elapsedTime = Time.time - startTime;
print("Elapsed time: " + elapsedTime);
startTime = Time.time;
int action = Mathf.FloorToInt(vectorAction[0]);
Point centerOfGravity = imageProcessor.CenterOfGravity;
@@ -55,15 +95,27 @@ public class CozmoAgent : Agent
{
case STOP:
movementController.currentMovementState = MovementState.Stop;
lastChosenMovement = MovementState.Stop;
//Test
SetReward(-0.1f);
break;
case FORWARD:
movementController.currentMovementState = MovementState.Forward;
lastChosenMovement = MovementState.Forward;
//Test
SetReward(0.01f);
break;
case RIGHT:
movementController.currentMovementState = MovementState.Right;
lastChosenMovement = MovementState.Right;
//Test
SetReward(-0.02f);
break;
case LEFT:
movementController.currentMovementState = MovementState.Left;
lastChosenMovement = MovementState.Left;
//Test
SetReward(-0.02f);
break;
default:
//movement.Move(0);
@@ -86,36 +138,43 @@ public class CozmoAgent : Agent
float reward = 0;
// Center of gravity is far away from the center (left)
if (centerOfGravityX <= centerOfImageX - nearAreaLimit)
if (centerOfGravityX <= centerOfImageX - nearAreaLimit && centerOfGravityX >= 0)
{
float range = centerOfImageX - nearAreaLimit;
reward = -(1 - (centerOfGravityX / range));
// Clamp the reward to max -1 in order to handle rewards if the center of gravity is outside of the image
reward = Mathf.Clamp(reward, -1, 0);
reward = Mathf.Clamp(reward, -1, 0) / 2;
}
// Center of gravity is near left of the center
else if (centerOfGravityX <= centerOfImageX)
else if ((centerOfGravityX <= centerOfImageX) && (centerOfGravityX >= (centerOfImageX - nearAreaLimit)))
{
float range = centerOfImageX - (centerOfImageX - nearAreaLimit);
float distanceToLeftFarBorder = centerOfGravityX - (centerOfImageX - nearAreaLimit);
reward = (distanceToLeftFarBorder / range);
}
// Center of gravity is far away from the center (right)
else if (centerOfGravityX >= centerOfImageX + nearAreaLimit)
else if ((centerOfGravityX >= (centerOfImageX + nearAreaLimit)) && (centerOfGravityX <= renderCamera.targetTexture.width))
{
float range = renderCamera.targetTexture.width - (centerOfImageX + nearAreaLimit);
reward = -(((centerOfGravityX - (centerOfImageX + nearAreaLimit)) / range));
// Clamp the reward to max -1 in order to handle rewards if the center of gravity is outside of the image
reward = Mathf.Clamp(reward, -1, 0);
reward = Mathf.Clamp(reward, -1, 0) / 2;
}
// Center of gravity is near right of the center
else if (centerOfGravityX >= centerOfImageX)
else if ((centerOfGravityX >= centerOfImageX) && (centerOfGravityX <= (centerOfImageX + nearAreaLimit)))
{
float range = (centerOfImageX + nearAreaLimit) - centerOfImageX;
float distanceToCenterOfImage = centerOfGravityX - centerOfImageX;
reward = (1 - distanceToCenterOfImage / range);
}
else
{
SetReward(-1);
AgentReset();
Debug.Log("Out of image range");
}
Debug.Log("Reward: " + reward);
SetReward(reward);
}