据我所知，你正在处理一个分配问题。您可以将scipy's linear_sum_assignment与距离矩阵一起用作成本矩阵。换句话说，您希望红色和蓝色点的组合产生较小的总距离（连接点之间的距离之和）。

import numpy as np 
import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import linear_sum_assignment

n_points = 5

np.random.seed(10)
blue_points = np.random.random((2, n_points))
red_points = np.random.random((2, n_points))

# Calculate distance matrix 
distance_matrix = np.linalg.norm(blue_points[:,:,np.newaxis] - red_points[:, np.newaxis,:], axis=0)

# Solve the assignment problem
blue_i, red_i = linear_sum_assignment(distance_matrix)

pairs = list(zip(blue_i, red_i))
# Plot 
fig, ax = plt.subplots()
ax.plot(*blue_points,'bo')
ax.plot(*red_points,'ro')
for i, j in pairs: 
    blue_points.shape
    x = [blue_points[0, i], red_points[0, j]]
    y = [blue_points[1, i], red_points[1, j]]
    ax.plot(x,y, 'k')