遗传算法最短路径matlab代码

下面是一个简单的遗传算法最短路径的 MATLAB 代码示例：

% 遗传算法最短路径示例
% 假设有一个5个城市的旅行商问题
% 假设城市坐标矩阵为cities，其中每一行表示一个城市的坐标
cities = [0.5, 1.5; 1, 2; 1.5, 1; 2, 0.5; 1, 1];

% 城市数量
nCities = size(cities, 1);

% 种群数量
nPop = 50;

% 迭代次数
nIter = 100;

% 交叉概率
pCrossover = 0.8;

% 变异概率
pMutation = 0.05;

% 初始化种群
pop = zeros(nPop, nCities);
for i = 1:nPop
    pop(i, :) = randperm(nCities);
end

% 迭代
for iter = 1:nIter
    % 评估种群中每个个体的适应度
    fitness = zeros(nPop, 1);
    for i = 1:nPop
        fitness(i) = getPathLength(cities, pop(i, :));
    end
    
    % 选择
    selectedPop = zeros(nPop, nCities);
    for i = 1:nPop
        % 轮盘赌选择
        idx1 = rouletteWheelSelection(fitness);
        idx2 = rouletteWheelSelection(fitness);
        % 选择两个不同的个体
        while idx2 == idx1
            idx2 = rouletteWheelSelection(fitness);
        end
        % 交叉
        if rand() < pCrossover
            [child1, child2] = crossover(pop(idx1, :), pop(idx2, :));
            selectedPop(i, :) = child1;
            % 有一定概率发生变异
            if rand() < pMutation
                selectedPop(i, :) = mutate(selectedPop(i, :));
            end
            i = i + 1;
            selectedPop(i, :) = child2;
            if rand() < pMutation
                selectedPop(i, :) = mutate(selectedPop(i, :));
            end
        else
            selectedPop(i, :) = pop(idx1, :);
        end
    end
    
    % 替换原种群
    pop = selectedPop;
end

% 找到最优解
minPath = inf;
for i = 1:nPop
    pathLength = getPathLength(cities, pop(i, :));
    if pathLength < minPath
        minPath = pathLength;
        bestPath = pop(i, :);
    end
end
% 输出结果
fprintf('最短路径长度: %f\n', minPath);
fprintf('最短路径: %s\n', mat2str(bestPath));

其中，`getPathLength` 函数计算路径长度，`rouletteWheelSelection` 函数进行轮盘赌选择，`crossover` 函数进行交叉操作，`mutate` 函数进行变异操作。这些函数可以根据具体需要进行编写。