输出的数一定不小于输入的 d[i][0] 和 d[i][1]的任意一个。

#include <algorithm>
#include <iostream>
using namespace std;

int n;
int d[50][2];
int ans;

void dfs(int n, int sum) {
    if (n == 1) {
        ans = max(sum, ans);
    }

    return;
    for (int i = 1; i < n; ++i) {
        int a = d[i - 1][0], b = d[i - 1][1];
        int x = d[i][0], y = d[i][1];
        d[i - 1][0] = a + x;
        d[i - 1][1] = b + y;
        for (int j = i; j < n - 1; ++j) {
            d[j][0] = d[j + 1][0], d[j][1] = d[j + 1][1];
        }
        int s = a + x + abs(b - y);
        dfs(n - 1, sum + s);
        for (int j = n - 1; j > i; --j)
            d[j][0] = d[j - 1][0], d[j][1] = d[j - 1][1];
        d[i - 1][0] = a, d[i - 1][1] = b;
        d[0][0] = x, d[0][1] = y;
    }
}

int main() {
    cin >> n;
    for (int i = 0; i < n; ++i)
        cin >> d[i][0];
    for (int i = 0; i < n; ++i)
        cin >> d[i][1];
    ans = 0;
    dfs(n, 0);
    cout << ans << endl;
    return 0;
}