如何统一处理随机产生的障碍物？

问题描述

我正在尝试处理对象生成器，但是它不能如我所愿。每当我的计时器通过对象的maxTime时，它都应该创建一个新对象，但是它只会执行一次并且在错误的位置（超出我的范围）。 这是我的代码：

public class Spawner : MonoBehavIoUr {

    public float maxTime = 10;
    private float timer = 0;
    public GameObject obstacle;
    private float width = (Screen.width)/2; //It doesn't recognize width as the variable that refeers the width of the screen,so i tried it
    private float maxTime0 = 10;
    public int pontos = 5;
    public float score;

    void Start() {
        GameObject new_obstacle = Instantiate(obstacle);
        new_obstacle.transform.position = transform.position + new Vector3(Random.Range(-width,width),0);
    }

    void Update()
    {
        if(timer > maxTime) {
            GameObject new_obstacle = Instantiate(obstacle);
            new_obstacle.transform.position = transform.position + new Vector3(Random.Range(-width,0);
            if(new_obstacle.transform.position.y < (Screen.height - 7)){
                DestroyImmediate(new_obstacle,true);
            }
            timer = 0; 
            maxTime = 0.9f * maxTime;
        }
        score += pontos; 
        timer += Time.deltaTime; 
    }
}

解决方法

听起来主要是因为您以 pixel 坐标生成了对象，因为Screen.width是 pixels 的屏幕宽度。

但是您更想要的是在Camera的平截头体中产生。

团结提供了一种找到Frustrum Size At Distance的方法。

因为您还要检查

   if(new_obstacle.transform.position.y < (Screen.height - 7))

要销毁您的物件，我敢打赌您的Update确实会产生该物件，但会立即销毁它们！

原因：您的生成者处于世界坐标，并且可能始终位于像素坐标的Screen.height - 7下！

由于您没有在Start中签入，这是它为您“工作”的唯一时间。

如果您真的需要此检查，则需要使用Camera.WorldToScreenPoint

将位置转换为屏幕像素空间。

Camera camera;

void Start() 
{
    camera = Camera.main;
    // assuming that your camera points exactly forward (2D game like)
    var distance = Mathf.Abs(transform.position.z - camera.tramsform.position.z);
    var frustumHeight = 2.0f * distance * Mathf.Tan(camera.fieldOfView * 0.5f * Mathf.Deg2Rad);
    var frustrumWidth = frustumHeight * camera.aspect;

    // not anymore in pixel space but frustrum width in world coordinates
    // at the z distance of the spawner
    width = frustrumWidth / 2f;


    var new_obstacle = Instantiate(obstacle);
    new_obstacle.transform.position = transform.position + Vector3.right * Random.Range(-width,width);
}

void Update()
{
    if(timer > maxTime) 
    {
        var new_obstacle = Instantiate(obstacle);
        new_obstacle.transform.position = transform.position + Vector3.right * Random.Range(-width,width);

        if(camera.WorldToScreenPoint(new_obstacle.transform.position).y < Screen.height - 7)
        {
            DestroyImmediate(new_obstacle,true);
        }
        
        timer = 0; 
        maxTime = 0.9f * maxTime;
    }

    // as mentioned here multiply by Time.deltaTime in order to
    // convert from pontos per frame into pontos per second
    score += pontos * Time.deltaTime; 
    timer += Time.deltaTime; 
}

是的，那么您可以使用Coroutine，它通常更易于维护..实际上是一种口味问题

void Start() 
{
    camera = Camera.main;
    // assuming that your camera points exactly forward (2D game like)
    var distance = Mathf.Abs(transform.position.z - camera.tramsform.position.z);
    var frustumHeight = 2.0f * distance * Mathf.Tan(camera.fieldOfView * 0.5f * Mathf.Deg2Rad);
    var frustrumWidth = frustumHeight * camera.aspect;

    // not anymore in pixel space but frustrum width in world coordinates
    // at the z distance of the spawner
    width = frustrumWidth / 2f;


    StartCoroutine(SpawnRoutine());
}

IEnumerator SpawnRoutine()
{
    // looks dangerous but is fine for Coroutines as long as you yield somewhere
    while(true)
    {
        var new_obstacle = Instantiate(obstacle);
        new_obstacle.transform.position = transform.position + Vector3.right * Random.Range(-width,true);
        }
       
        // In Coroutines a yield tells Unity to "pause" here,// render this frame and continue from here in the next frame
        yield return new WaitForSeconds(maxTime);

        maxTime = 0.9f * maxTime;
    }
}

private void Update()
{
    score += pontos * Time.deltaTime; 
}