LeetCode1197_进击的骑士_BFS_贪心

最新推荐文章于 2024-06-24 20:29:36 发布

驹子还是叶子

最新推荐文章于 2024-06-24 20:29:36 发布

阅读量131

点赞数 1

文章标签：宽度优先算法

本文链接：https://blog.csdn.net/qq_40992456/article/details/130858245

版权

视频讲解版：LeetCode1197_进击的骑士_BFS_贪心

class Solution {
    private int[][] directions = {{1, 2}, {2, 1}, {2, -1}, {1, -2}, {-1, -2}, {-2, -1}, {-2, 1}, {-1, 2}};
    public int minKnightMoves(int x, int y) {
        if (x == 0 && y == 0) {
            return 0;
        }
        // 访问数组：boolean[][] visited(二、三、四象限有负值，需特殊处理)
        boolean[][] visited = new boolean[601][601];
        Deque<int[]> deque = new LinkedList<>();
        deque.offer(new int[]{0, 0});
        visited[300][300] = true;
        int step = 0;
        while (!deque.isEmpty()) {
            step++;
            int size = deque.size();
            for (int i = 0; i < size; i++) {
                int[] cur = deque.poll();
                // 当前位置到目标位置的曼哈顿距离
                /**
                    额外小知识：曼哈顿距离
                    例如在平面上，坐标(x1,y1)的i点与坐标(x2,y2)的j点的曼哈顿距离为：
                    d(i,j)=|X1-X2|+|Y1-Y2|.
                 */
                int currentDistance = getDistance(cur[0], cur[1], x, y);
                for (int[] direction : directions) {
                    int rowIndex = cur[0] + direction[0];
                    int colIndex = cur[1] + direction[1];
                    if (rowIndex == x && colIndex == y) {
                        return step;
                    }
                    if (visited[rowIndex + 300][colIndex + 300]) {
                        continue;
                    }
                    // 新到达的位置到目标位置的曼哈顿距离 小于 当前位置到目标位置的曼哈顿距离
                    int updateDistance = getDistance(rowIndex, colIndex, x, y);
                    if (updateDistance < currentDistance || step == 1) {
                        deque.offer(new int[]{rowIndex, colIndex});
                        visited[rowIndex + 300][colIndex + 300] = true;
                    }
                }
            }
        }
        return -1;
    }

    private int getDistance(int srcX, int srcY, int targetX, int targetY) {
        return Math.abs(srcX - targetX) + Math.abs(srcY - targetY);
    }
}