public override void AgentAction(float[] act, string textAction)
{
if (agentDone)
{
if (!cr_.allDone_)
{
return;
}
else
{
agentDone = false;
cr_.setAgentActiveStatus(sw_id, true);
return;
//Done();
}
}
//Debug.Log("Actions act[0] " + act[0] + "act[1] " + act[1]);
step++;
// 0 -> move forward
// 1 -> turn left
// 2 -> turn right
float distToTargetOld = (target_.transform.position - agent_.pos).magnitude;
float orientationOld = agent_.cosineOrientation();
if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
{
gameObject.transform.position += gameObject.transform.forward.normalized * Mathf.Clamp(act[0], 0f, 1f);
float rotAngle = 90.0f * Mathf.Clamp(act[1], -1f, 1f); // opposite ends, reachable from [-90, 90] rotation
gameObject.transform.Rotate(0f, rotAngle, 0f);
}
else
{
int action = (int)act[0];
//Debug.Log(action);
if (action == 0) // walk forward
{
agent_.walkForward();
}
if (action == 1) // turn right
{
agent_.turnRight();
}
else if (action == 2) // turn left
{
agent_.turnLeft();
}
else if (action == 3) // stop
{
agent_.Brake();
}
gameObject.transform.position = agent_.pos;
gameObject.transform.forward = agent_.forward;
}
cr_.setAgent(ref agent_, sw_id);
float distToTarget = (target_.transform.position - agent_.pos).magnitude;
float orientation = agent_.cosineOrientation();
if (agent_.targetReached())
{
//Debug.Log("Reached Target! Agent Done.");
AddReward(rewardTargetReached);
AgentDoneStuff();
return;
}
if (!agent_.withinBounds(cr_.areaMinBound_, cr_.areaMaxBound_) || step > maxSteps)
{
//Debug.Log("Went out of Arena! Agent Done.");
AddReward(rewardOutOfBounds);
AgentDoneStuff();
return;
}
if (cr_.doesCollide(sw_id))
{
Debug.Log("Collision!");
AddReward(rewardCollision);
}
//reward for gaining distance towards the target
AddReward(distanceGainedWeight * (distToTargetOld - distToTarget)); // should have come closer to target
//reward for orienting towards the target
AddReward(orientationGainedWeight * (orientation - orientationOld)); // should have aligned better with target
//reward for each step (usually negative)
AddReward(rewardEachStep);
}