New Reward function and MovementController

This commit is contained in:
Tobi
2019-05-23 17:28:37 +02:00
parent 411338a35d
commit 221ffb54dc
5 changed files with 156 additions and 193 deletions
+57 -26
View File
@@ -19,32 +19,32 @@ public class CozmoAgent : Agent
[Tooltip("The virtual Cozmo camera")]
public Camera renderCamera;
//[Tooltip("Final cropped and scaled rendertexture")]
//public RenderTexture renderTextureScaled;
[Tooltip("Reference to the CozmoMovement script")]
public CozmoMovement movement;
public CozmoMovementController movementController;
public float timeBetweenDecisionsAtInference;
private Academy academy; // CozmoAcademy
private float timeSinceDecision; // time since last decision
private ImageProcessor onRenderImageTest; // reference to the ImageProcessor
private ImageProcessor imageProcessor; // reference to the ImageProcessor
private int nearAreaLimit = 0; // X coordinate limit for the near to the imagecenter area
private int farAreaLimit = 0; // X coordinate limit for the far away to the imagecenter area
// for testing
//private float[] floats = { 1.0f, 2.0f, 3.0f };
private void Start()
{
academy = FindObjectOfType(typeof(CozmoAcademy)) as CozmoAcademy;
onRenderImageTest = renderCamera.GetComponent<ImageProcessor>();
imageProcessor = renderCamera.GetComponent<ImageProcessor>();
nearAreaLimit = (int)(renderCamera.targetTexture.width / 2 * NEAR_AREA_PERCENTAGE_OFFSET);
farAreaLimit = (int)(renderCamera.targetTexture.width / 2 * FAR_AREA_PERCENTAGE_OFFSET);
}
public void FixedUpdate()
{
WaitTimeInference();
// for testing
//AgentAction(floats, "ActionText");
}
@@ -53,7 +53,7 @@ public class CozmoAgent : Agent
{
//print("Action before FloorToInt: " + vectorAction[0]);
int action = Mathf.FloorToInt(vectorAction[0]);
Point centerOfGravity = onRenderImageTest.CenterOfGravity;
Point centerOfGravity = imageProcessor.CenterOfGravity;
//Vector3 targetPos = transform.position;
//print("Action after FloorToInt: " + action);
@@ -63,16 +63,16 @@ public class CozmoAgent : Agent
switch (action)
{
case STOP:
movement.Move(0);
movementController.currentMovementState = MovementState.Stop;
break;
case FORWARD:
movement.Move(1);
movementController.currentMovementState = MovementState.Forward;
break;
case RIGHT:
movement.Turn(1);
movementController.currentMovementState = MovementState.Right;
break;
case LEFT:
movement.Turn(-1);
movementController.currentMovementState = MovementState.Left;
break;
default:
//movement.Move(0);
@@ -83,28 +83,61 @@ public class CozmoAgent : Agent
if (renderCamera != null)
{
renderCamera.Render();
//onRenderImageTest.ProcessRenderTexture_NEW(renderCamera.targetTexture, onRenderImageTest.renderTextureCropped);
}
RewardAgent();
imageProcessor.enabled = false;
}
// If centerOfGravity lies near to the center of the image horizontally
if (centerOfGravity.X > renderCamera.targetTexture.width / 2 - nearAreaLimit && centerOfGravity.X < renderCamera.targetTexture.width / 2 + nearAreaLimit)
/// <summary>
/// TODO: Cleanup code
/// </summary>
private void RewardAgent()
{
float centerOfImageX = renderCamera.targetTexture.width / 2;
float centerOfGravityX = imageProcessor.CenterOfGravity.X;
float reward = 0;
// Center of gravity is far left of the center
if (centerOfGravityX <= centerOfImageX - farAreaLimit)
{
SetReward(1);
print("Reward: +1");
reward = -1;
}
else if (centerOfGravity.X > renderCamera.targetTexture.width / 2 - farAreaLimit && centerOfGravity.X < renderCamera.targetTexture.width / 2 + farAreaLimit)
// Center of gravity is between far and near left of the center
else if (centerOfGravityX <= centerOfImageX - nearAreaLimit)
{
SetReward(-0.3f);
print("Reward: -0.3");
float range = (centerOfImageX - nearAreaLimit) - (centerOfImageX - farAreaLimit);
float distanceToLeftFarBorder = centerOfGravityX - (centerOfImageX - farAreaLimit);
reward = -(1 - (distanceToLeftFarBorder / range));
}
else
// Center of gravity is near left of the center
else if (centerOfGravityX <= centerOfImageX)
{
SetReward(-1);
print("Reward: -1");
float range = centerOfImageX - (centerOfImageX - nearAreaLimit);
float distanceToLeftFarBorder = centerOfGravityX - (centerOfImageX - nearAreaLimit);
reward = (distanceToLeftFarBorder / range);
}
// Center of gravity is far right of the center
else if (centerOfGravityX >= centerOfImageX + farAreaLimit)
{
reward = -1;
}
// Center of gravity is between far and near right of the center
else if (centerOfGravityX >= centerOfImageX + nearAreaLimit)
{
float range = (centerOfImageX + farAreaLimit) - (centerOfImageX + nearAreaLimit);
float distanceToLeftFarBorder = centerOfGravityX - (centerOfImageX + nearAreaLimit);
reward = -(distanceToLeftFarBorder / range);
}
// Center of gravity is near right of the center
else if (centerOfGravityX >= centerOfImageX)
{
float range = (centerOfImageX + nearAreaLimit) - centerOfImageX;
float distanceToLeftFarBorder = centerOfGravityX - centerOfImageX;
reward = (1 - distanceToLeftFarBorder / range);
}
onRenderImageTest.enabled = false;
SetReward(reward);
}
// to be implemented by the developer
@@ -132,7 +165,6 @@ public class CozmoAgent : Agent
if (!academy.GetIsInference())
{
//onRenderImageTest.ProcessRenderTexture_NEW(renderCamera.targetTexture, onRenderImageTest.renderTextureCropped);
RequestDecision();
}
else
@@ -140,7 +172,6 @@ public class CozmoAgent : Agent
if (timeSinceDecision >= timeBetweenDecisionsAtInference)
{
timeSinceDecision = 0f;
//onRenderImageTest.ProcessRenderTexture_NEW(renderCamera.targetTexture, onRenderImageTest.renderTextureCropped);
RequestDecision();
}
else