关于“js+ajax实现的A*游戏路径算法整理”的完整攻略，以下是详细介绍（注意，为了方便阅读，带有代码块的内容使用了代码语法高亮）：
A*算法是一种基于图形、搜索和启发式运算的寻路算法，通常用于从起点到目标点的最优路径搜索。
A*算法将费用（距离、代价）与启发式函数两者结合，来评估当前节点到目标点路径的可能代价大小。其中启发式函数具有预测未来路径代价的功能。
1.置入起始点。
2.置起始点的f=0，算法对已处理的部分为0，未处理部分为无限制。
3.若终止则结束。
4.选取“未处理”状态中f值最小的点做为当前节点。
5.将当前节点从“未处理”中删除，放入“已处理”中。
6.扩展当前节点。
7.若候选节点中已存在于已处理表，或未处理表f值比当前节点要小，则删除该节点。
8.将本次扩展得到的所有节点按f值排序后，加入未处理节点。
下面是一段基于A*算法的JavaScript代码，实现了地图寻路的功能，具体实现过程详见注释：
// 定义地图类
class Map {
constructor() {
this.map = mapData; // 初始化地图数据
this.start = null; // 起点
this.end = null; // 终点
}
// 计算当前点到终点的曼哈顿距离
calcManhattan(point) {
const dx = Math.abs(point.x - this.end.x);
const dy = Math.abs(point.y - this.end.y);
return dx + dy;
}
// 查找当前点的八个相邻节点
findNeighbor(point) {
const w = this.map[0].length;
const h = this.map.length;
const neighbor = [];
for (let i = point.x - 1; i <= point.x + 1; i++) {
for (let j = point.y - 1; j <= point.y + 1; j++) {
if (i >= 0 && i < h && j >= 0 && j < w && (i != point.x || j != point.y) && this.map[i][j] != 1) {
neighbor.push({x: i, y: j});
}
}
}
return neighbor;
}
//寻路
findPath() {
let openList = [];
let closedList = [];
openList.push(this.start); // 起点放入openList
while (openList.length) { // 当openList不为空时
let minIndex = 0;
// 找到f(n)=g(n)+h(n)最小的节点作为当前节点
for (let i = 0; i < openList.length; i++) {
if (openList[i].f < openList[minIndex].f) {
minIndex = i;
}
}
const currentNode = openList[minIndex]; // 当前节点
// 当前节点为终点时
if (currentNode.x === this.end.x && currentNode.y === this.end.y) {
let path = [];
let node = currentNode;
while (node) {
path.push(node);
node = node.parent;
}
return path.reverse();
}
// 将当前点从`openList`转移到`closedList`
openList.splice(minIndex, 1);
closedList.push(currentNode);
// 扩展相邻节点
let neighbors = this.findNeighbor(currentNode);
for (let i = 0; i < neighbors.length; i++) {
const neighbor = neighbors[i];
// 如果邻居节点已被访问过或被加入到开启列表中，则继续下一个节点的遍历
if (closedList.some(v => v.x === neighbor.x && v.y === neighbor.y)
|| openList.some(v => v.x === neighbor.x && v.y === neighbor.y)) {
continue;
}
let g = currentNode.g + 1; // 到达邻居节点的实际代价
// 确认邻居节点代价是否最优
let isInOpenList = openList.some(v => v.x === neighbor.x && v.y === neighbor.y);
let isInClosedList = closedList.some(v => v.x === neighbor.x && v.y === neighbor.y);
let h = this.calcManhattan(neighbor); // 当前节点到终点的曼哈顿距离
let f = g + h; // 总代价
if (!isInOpenList && !isInClosedList) {
openList.push({
x: neighbor.x,
y: neighbor.y,
g,
f,
parent: currentNode
});
} else if (isInOpenList && g < neighbor.g) {
const index = openList.findIndex(v => v.x === neighbor.x && v.y === neighbor.y);
openList[index].g = g;
openList[index].f = f;
openList[index].parent = currentNode;
} else if (isInClosedList && g < neighbor.g) {
const index = closedList.findIndex(v => v.x === neighbor.x && v.y === neighbor.y);
closedList[index].g = g;
closedList[index].f = f;
closedList[index].parent = currentNode;
openList.push(closedList[index]);
closedList.splice(index, 1);
}
}
}
return null;
}
}
// 示例#1的应用
const mapData = [
[0, 0, 0, 0, 0],
[0, 0, 0, 1, 0],
[0, 0, 0, 1, 0],
[0, 0, 0, 0, 0],
];
const map = new Map();
map.start = { x: 0, y: 0 }; // 起点
map.end = { x: 3, y: 4 }; // 终点
console.log(map.findPath());
下面演示如何用A*算法实现通用的寻路：
class PathFinding {
constructor() {
// 地图大小
this._size = {
x: 0,
y: 0
};
// 障碍物数组
this._barriers = [];
}
// 初始化地图和障碍物数组
init(size, barriers) {
this._size = size;
this._barriers = barriers;
}
//判断是否是障碍物
isBarrier(x, y) {
let isBarrier = false;
for (let i = 0; i < this._barriers.length; i++) {
if (this._barriers[i].x === x && this._barriers[i].y === y) {
isBarrier = true;
break;
}
}
return isBarrier;
}
// 计算曼哈顿距离
calcManhattan(from, to) {
return Math.abs(to.x - from.x) + Math.abs(to.y - from.y);
}
// 查找路径
findPath(src, dest) {
let openList = [];
let closedList = [];
let startNode = {
x: src.x,
y: src.y,
g: 0, // 实际代价
h: this.calcManhattan(src, dest), // 估测代价
f: this.calcManhattan(src, dest), // 总代价
parent: null
};
openList.push(startNode);
while (openList.length > 0) { // 当openList不为空时
// 从openList中取出f(n)值最小的作为当前节点
let currentIndex = 0;
for (let i = 1; i < openList.length; i++) {
if (openList[i].f < openList[currentIndex].f) {
currentIndex = i;
}
}
let currentNode = openList[currentIndex];
// 如果找到目标，则返回路径
if (currentNode.x === dest.x && currentNode.y === dest.y) {
let path = [];
let node = currentNode;
while (node) {
path.push(node);
node = node.parent;
}
return path.reverse();
}
// 从openList删除当前节点，加入到closedList
openList.splice(currentIndex, 1);
closedList.push(currentNode);
// 扩展节点
for (let i = currentNode.x - 1; i <= currentNode.x + 1; i++) {
for (let j = currentNode.y - 1; j <= currentNode.y + 1; j++) {
if (i >= 0 && i < this._size.x && j >= 0 && j < this._size.y
&& (i != currentNode.x || j != currentNode.y) && !this.isBarrier(i, j)) {
let g = currentNode.g + 1; // 实际代价
let h = this.calcManhattan({ x: i, y: j }, dest); // 估测代价
let f = g + h; // 总代价
let isExistInClosedList = false;
for (let k = 0; k < closedList.length; k++) {
if (closedList[k].x === i && closedList[k].y === j) {
isExistInClosedList = true;
break;
}
}
let isExistInOpenList = false;
for (let k = 0; k < openList.length; k++) {
if (openList[k].x === i && openList[k].y === j) {
isExistInOpenList = true;
break;
}
}
if (!isExistInClosedList && !isExistInOpenList) {
openList.push({
x: i,
y: j,
g,
h,
f,
parent: currentNode
})
} else if (isExistInOpenList && g < openList[currentIndex].g) {
let index = openList.findIndex((value) => value.x === i && value.y === j);
openList[index].g = g;
openList[index].h = h;
openList[index].f = f;
openList[index].parent = currentNode;
}
}
}
}
}
}
};
// 示例#2的应用
// 定义地图大小和障碍物数组
let size = {x: 6, y: 3};
let barriers = [
{x: 1, y: 0},
{x: 1, y: 1},
{x: 3, y: 0},
{x: 3, y: 1},
{x: 4, y: 1},
{x: 5, y: 1},
];
// 初始化并进行寻路
const pf = new PathFinding();
pf.init(size, barriers);
console.log(pf.findPath({x: 0, y: 0}, {x: 5, y: 2}));
以上是关于js+ajax实现的A*游戏路径算法整理的详细攻略，希望能够帮到你！
本文链接：http://task.lmcjl.com/news/11837.html