AgentScript.cs

using System.Collections;
using System.Collections.Generic;
using Unity.MLAgents;
using Unity.MLAgents.Actuators;
using Unity.MLAgents.Sensors;
using UnityEngine;

public class AgentScript : Agent
{
    private Transform Target;
    void Awake()
    {

    }

    void Start()
    {
        Target = transform.parent.Find("PresentBoxPrefab");// end of arena transform
    }
    
    public void OnEpisodeEnd()
    {
        EndEpisode();
    }

    public override void CollectObservations(VectorSensor sensor)
    {
        // Our position and target position are observed
        sensor.AddObservation(transform.localPosition);
        sensor.AddObservation(Target.localPosition);
    }
    
    public override void OnActionReceived(ActionBuffers actions)
    {
        AddReward(-1f / MaxStep);
    }

    //public override void Heuristic(in ActionBuffers actionsOut)//do i still need to use Heuristic function when using InputActuator component?
    //{
    //}

    public void EndOfArena()
    {
        AddReward(1f);
        EndEpisode();
    }

    public void AgentDied()
    {
        AddReward(-0.5f);
        EndEpisode();
    }
}