-
Notifications
You must be signed in to change notification settings - Fork 8
/
Copy pathplot_result.py
155 lines (123 loc) · 4.45 KB
/
plot_result.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
import numpy as np
import matplotlib.pyplot as plt
import sys
import pymap3d as p3d
import pandas as pd
try:
path = sys.argv[1]
except:
path = "result/klt1_predict"
try:
title = sys.argv[2]+" "
except:
title = ""
item = ['gt','go','rtk','b','w','bw']
labels = ['goGPS','RTKLIB','TDL-B','TDL-W','Ground Truth','TDL-BW']
colors = ['orange', 'brown', 'blue', 'purple','green', 'red']
labels1 = ['goGPS','RTKLIB','TDL-B','TDL-W','TDL-BW']
colors1 = ['orange', 'brown', 'blue', 'purple', 'red']
gtpos = np.loadtxt(path+"/gt.csv",skiprows=1,delimiter=',')
gopos = np.loadtxt(path+"/gogps_pos.csv",skiprows=1,delimiter=',')
rtkpos = np.loadtxt(path+"/rtklib_pos.csv",skiprows=1,delimiter=',')
bpos = np.loadtxt(path+"/TDL_bias_pos.csv",skiprows=1,delimiter=',')
wpos = np.loadtxt(path+"/TDL_weight_pos.csv",skiprows=1,delimiter=',')
bwpos = np.loadtxt(path+"/TDL_bw_pos.csv",skiprows=1,delimiter=',')
gt_init = gopos[0]
errors = []
traj = []
for gt,go,rtk,b,w,bw in zip(gtpos,gopos,rtkpos,bpos,wpos,bwpos):
errors.append([
p3d.geodetic2enu(*go,*gt),
p3d.geodetic2enu(*rtk,*gt),
p3d.geodetic2enu(*b,*gt),
p3d.geodetic2enu(*w,*gt),
p3d.geodetic2enu(*bw,*gt)
])
traj.append([
p3d.geodetic2enu(*go,*gt_init),
p3d.geodetic2enu(*rtk,*gt_init),
p3d.geodetic2enu(*b,*gt_init),
p3d.geodetic2enu(*w,*gt_init),
p3d.geodetic2enu(*gt,*gt_init),
p3d.geodetic2enu(*bw,*gt_init)
])
errors = np.array(errors)
traj = np.array(traj)
print("2D error: ",np.linalg.norm(errors[:,:,:2],axis=2).mean(axis=0))
print("3D error: ",np.linalg.norm(errors,axis=2).mean(axis=0))
plt.rcParams.update({'font.size': 30, 'font.family': 'Times New Roman'})
errors3d = {}
# plt.figure(figsize=(16, 10))
# for i in range(5):
# plt.plot(np.linalg.norm(errors[:,i,:],axis=1),label = labels1[i], color=colors1[i])
# errors3d[labels1[i]] = np.linalg.norm(errors[:,i,:],axis=1)
# plt.xlim(left=0)
# plt.legend()
# plt.title(title+"3D Error(m)")
# plt.ylabel("Error(m)")
# plt.tight_layout()
# plt.savefig(path+"/error3d.png")
fig, (ax, ax2) = plt.subplots(2, 1, sharex=True, figsize=(16, 10))
# 处理每个标签和对应的颜色
errors3d = {}
for i in range(5):
norm_data = np.linalg.norm(errors[:, i, :], axis=1)
errors3d[labels1[i]] = norm_data
# 绘制到子图
ax.plot(norm_data, label=labels1[i], color=colors1[i])
ax2.plot(norm_data, label=labels1[i], color=colors1[i])
# 设置Y轴范围
ax.set_ylim(200, 1200) # 上图显示较大的错误值
ax2.set_ylim(0, 200) # 下图显示较小的错误值
# 隐藏两个子图间的间隙
ax.spines['bottom'].set_visible(False)
ax2.spines['top'].set_visible(False)
ax.xaxis.tick_top()
ax.tick_params(labeltop=False)
ax2.xaxis.tick_bottom()
# 断裂标记参数
d = 0.015 # 斜杠大小
kwargs = dict(transform=ax.transAxes, color='k', clip_on=False)
ax.plot((-d, +d), (-d, +d), **kwargs)
ax.plot((1 - d, 1 + d), (-d, +d), **kwargs)
kwargs.update(transform=ax2.transAxes)
ax2.plot((-d, +d), (1 - d, 1 + d), **kwargs)
ax2.plot((1 - d, 1 + d), (1 - d, 1 + d), **kwargs)
# 其他设置
ax.set_title(title + " 3D Error (m)")
ax.set_ylabel("Error (m)")
ax.legend()
plt.tight_layout()
# 保存图片
plt.savefig(path + "/error3d.png")
plt.show()
errors3d = pd.DataFrame(errors3d)
errors3d.to_csv(path+"/errors3d.csv",index=None)
# Redraw the boxplot with Times New Roman font
plt.figure(figsize=(12, 6))
plt.boxplot([errors3d['goGPS'], errors3d['RTKLIB'], errors3d['TDL-B'], errors3d['TDL-W'], errors3d['TDL-BW']],
labels=['goGPS', 'RTKLIB', 'TDL-B', 'TDL-W', 'TDL-BW'])
plt.title(f'Boxplot of 3D Positioning Errors on {title} Dataset')
plt.ylabel('Error (meters)')
plt.grid(True)
plt.savefig(path+"/error3dbox.png")
plt.figure(figsize=(16, 10))
for i in range(5):
plt.plot(np.linalg.norm(errors[:,i,:2],axis=1),label = labels1[i], color=colors1[i])
plt.xlim(left=0)
plt.legend()
plt.title(title+"2D Error(m)")
plt.ylabel("Error(m)")
plt.tight_layout()
plt.savefig(path+"/error2d.png")
plt.figure(figsize=(10, 10))
for i in range(6):
plt.plot(traj[:,i,0],traj[:,i,1], marker='o', linestyle='-', color=colors[i],label=labels[i],markersize=5)
handles, labels = plt.gca().get_legend_handles_labels()
order = [4, 0, 1, 2,3,5] # 指定图例的顺序：line3, line1, line2
plt.legend([handles[idx] for idx in order], [labels[idx] for idx in order])
plt.title(title+"2D Trajectory(m)")
plt.axis('equal')
plt.tight_layout()
plt.savefig(path+"/traj2d.png")
plt.show()