3D游戏编程实践——P&D 过河游戏智能帮助实现

本文详细介绍了使用C++实现P&D过河游戏的智能算法,包括状态图自动生成、图数据表示方法及最短路径算法SPFA的应用。通过构建状态转移矩阵,实现了游戏中的最优路径计算。
部署运行你感兴趣的模型镜像

P&D 过河游戏智能帮助实现

需求

  • 实现状态图的自动生成
  • 讲解图数据在程序中的表示方法
  • 利用算法实现下一步的计算

实现过程

实现状态图的自动生成&讲解图数据在程序中的表示方法

牧师与魔鬼状态可以用一个三元组(O,P,D)来表示,分别表示当前船的位置(0表示在出发地,1表示在目的地),出发地牧师数量和出发地魔鬼数量,状态数一共是 2 ∗ 4 ∗ 4 = 32 2 * 4*4=32 244=32

我先用c++写了一个小程序构造出状态转移矩阵

首先求出每个状态下转移的邻接矩阵,转移表存储在数组 a [ 32 ] [ 32 ] a[32][32] a[32][32]中,如果状态i可以一步到达状态j,则 a [ i ] [ j ] = 1 a[i][j]=1 a[i][j]=1,否则 a [ i ] [ j ] = 0 a[i][j]=0 a[i][j]=0

int get(int o,int i,int j){ //将三元组转化成状态编号
    return o*16+i*4+j;
}
bool jud(int o,int i,int j){ //判断当前状态是否合法
    if (i<0||i>3||j<0||j>3) return 0;
    if ((i<j)&&(i!=0)) return 0;
    if ((3-i<3-j)&&(3-i!=0)) return 0;
    return 1;
}
#define rep(i,l,r) for (int i=l;i<=r;i++)
rep(o,0,1){
    rep(i,0,3){
        rep(j,0,3){
            int st=get(o,i,j);
            if (!jud(o,i,j)) continue;
            if (o==0){  //from 出发地的牧师和魔鬼只可能减少
                rep(dx,-2,0) rep(dy,-2,0) if (1<=abs(dx+dy)&&abs(dx+dy)<=2){
                    if (!jud(o^1,i+dx,j+dy)) continue;
                    a[get(o,i,j)][get(o^1,i+dx,j+dy)]=1;
                }
            }
            else { //to 出发地的牧师和魔鬼只可能增加
                rep(dx,0,2) rep(dy,0,2) if (1<=dx+dy&&dx+dy<=2){
                    if (!jud(o^1,i+dx,j+dy)) continue;
                    a[get(o,i,j)][get(o^1,i+dx,j+dy)]=1;
                }
            }
        }
    }
}

然后用最短路算法spfa求出所有状态到终点(1,0,0)的最短路距离

rep(o,0,1) rep(i,0,3) rep(j,0,3) dis[get(o,i,j)]=inf; 
dis[get(1,0,0)]=0;
queue<int> q; q.push(get(1,0,0)); vis[get(1,0,0)]=1; 
while (!q.empty()){
    int u=q.front(); q.pop(); vis[u]=0;
    int uo=u/16,ux=(u%16)/4,uy=u%4;
    rep(i,-2,2) rep(j,-2,2) {
        int vo=uo^1,vx=ux+i,vy=uy+j;
        int v=get(vo,ux+i,uy+j);
        if (jud(vo,vx,vy)&&a[v][u]&&dis[v]>dis[u]+1){  
            //如果状态v能到达状态u且能够更新状态v的距离
            dis[v]=dis[u]+1;
            if (!vis[v]) vis[v]=1,q.push(v);
        }
    }
}

我们只需要保留在最短路图上的边并输出

rep(o,0,1) rep(i,0,3) rep(j,0,3){
    rep(x,0,3) rep(y,0,3) {
        if (dis[get(o,x,y)]==dis[get(o^1,i,j)]+1&&a[get(o,x,y)][get(o^1,i,j)]) a[get(o,x,y)][get(o^1,i,j)]=1;
        else a[get(o,x,y)][get(o^1,i,j)]=0;
    }
}
rep(o,0,1) rep(i,0,3) rep(j,0,3){
    printf("{");
    rep(no,0,1) rep(x,0,3) rep(y,0,3) {
        printf("%d",a[get(o,i,j)][get(no,x,y)]);
        if (x==3&&y==3&&no==1) printf("},\n");
        else printf(",");
    }
}
状态转移矩阵
{{0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,1,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0},
            {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0}}

在游戏中通过这张状态表就可以得知下一步应该到达哪一个状态了。

利用算法实现下一步的计算

在之前作业5中(https://blog.youkuaiyun.com/ctlchild/article/details/101454426

加入类AI,其中实现了函数tip,读取起点终点的牧师魔鬼数量,船的状态,计算出当前的状态,并根据转移表选择下一步的选择。

public int tip(CoastCon fromCoast,CoastCon toCoast,BoatCon boat) {	
    int fromP = 0, fromD = 0, toP = 0, toD = 0, boatst = 0;
    int[] fromCount = fromCoast.getCharacterNum();
    fromP += fromCount[0];
    fromD += fromCount[1];

    int[] toCount = toCoast.getCharacterNum ();
    toP += toCount[0];
    toD += toCount[1];

    int[] boatCount = boat.getCharacterNum ();
    if (boat.getStatus ()==-1) boatst=1;
    else boatst=0;
    if (boat.getStatus () == -1) {
        toP += boatCount[0]; toD += boatCount[1];	// boat at toCoast
    }
    else {
        fromP += boatCount[0]; fromD += boatCount[1];	// boat at fromCoast	
    }
    Debug.Log(fromP.ToString()+' '+fromD.ToString()+' '+get(boatst,fromP,fromD).ToString());
    for (int o=0;o<=1;o++)
        for (int i=0;i<=3;i++) 
            for (int j=0;j<=3;j++) 
                if (trans[get(boatst,fromP,fromD),get(o,i,j)]==1) return get(boatst,i,j);

    return -1;
}

运行效果

在这里插入图片描述

您可能感兴趣的与本文相关的镜像

Stable-Diffusion-3.5

Stable-Diffusion-3.5

图片生成
Stable-Diffusion

Stable Diffusion 3.5 (SD 3.5) 是由 Stability AI 推出的新一代文本到图像生成模型,相比 3.0 版本,它提升了图像质量、运行速度和硬件效率

【评估多目标跟踪方法】9个高度敏捷目标在编队中的轨迹和测量研究(Matlab代码实现)内容概要:本文围绕“评估多目标跟踪方法”,重点研究9个高度敏捷目标在编队飞行中的轨迹生成与测量过程,并提供完整的Matlab代码实现。文中详细模拟了目标的动态行为、运动约束及编队结构,通过仿真获取目标的状态信息与观测数据,用于验证和比较不同多目标跟踪算法的性能。研究内容涵盖轨迹建模、噪声处理、传感器测量模拟以及数据可视化等关键技术环节,旨在为雷达、无人机编队、自动驾驶等领域的多目标跟踪系统提供可复现的测试基准。; 适合人群:具备一定Matlab编程基础,从事控制工程、自动化、航空航天、智能交通或人工智能等相关领域的研究生、科研人员及工程技术人员。; 使用场景及目标:①用于多目标跟踪算法(如卡尔曼滤波、粒子滤波、GM-CPHD等)的性能评估与对比实验;②作为无人机编队、空中交通监控等应用场景下的轨迹仿真与传感器数据分析的教学与研究平台;③支持对高度机动目标在复杂编队下的可观测性与跟踪精度进行深入分析。; 阅读建议:建议读者结合提供的Matlab代码进行实践操作,重点关注轨迹生成逻辑与测量模型构建部分,可通过修改目标数量、运动参数或噪声水平来拓展实验场景,进一步提升对多目标跟踪系统设计与评估的理解。
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
红包 添加红包
表情包 插入表情
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值