myRFS_code/Data_generator.py at My_Bayesian_lab · 326037785/myRFS_code · GitHub

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
from matplotlib import pyplot as plt
import numpy as np
import time
import scipy.stats
from numpy import newaxis as na
import joblib
from scipy.optimize import linear_sum_assignment
class Model:
    def __init__(self,dt=1):
        self.dt  = dt
        self.x_dim = 4
        self.z_dim = 2
        self.F = np.array([[1,dt,0,0],
                           [0,1,0,0],
                           [0,0,1,dt],
                           [0,0,0,1]])
        self.B = np.array([[dt**2/2, 0],
                           [dt, 0],
                           [0, dt**2/2],
                           [0, dt]])
        self.sigmaV = 1
        self.Q = self.sigmaV**2*self.B.dot(self.B.T)
        R = np.asarray(np.diag([2,2]))
        self.R = np.multiply(R,R)
        self.H = np.array([[1,0,0,0],
                          [0,0,1,0]])
        self.P_S = 0.99
        self.Q_S = 1-self.P_S
        self.P_D = 0.68
        self.Q_D = 1-self.P_D

        ## possion birth RFS
        self.T_birth = 4
        self.L_birth = 4
        self.r_birth = 0.1*np.ones((self.L_birth,))
        self.w_birth = 0.03*np.ones((self.L_birth,))
        self.m_birth = np.zeros((self.x_dim,self.L_birth))
        self.B_birth = np.zeros((self.x_dim,self.x_dim,self.L_birth))
        self.P_birth = np.zeros((self.x_dim,self.x_dim,self.L_birth))

        self.m_birth[:,0] = np.array([0, 0, 0, 0]).astype(np.float)
        self.B_birth[...,0] = np.diag([10,10,10,10]).astype(np.float)
        self.P_birth[...,0] = self.B_birth[...,0].dot(self.B_birth[...,0].T)

        self.m_birth[:,1] = np.array([400, 0, -600, 0]).astype(np.float)
        self.B_birth[...,1] = np.diag([10,10,10,10]).astype(np.float)
        self.P_birth[...,1] = self.B_birth[...,1].dot(self.B_birth[...,1].T)

        self.m_birth[:,2] = np.array([-500, 0, -400, 0]).astype(np.float)
        self.B_birth[...,2] = np.diag([10,10,10,10]).astype(np.float)
        self.P_birth[...,2] = self.B_birth[...,2].dot(self.B_birth[...,2].T)

        self.m_birth[:,3] = np.array([-300, 0, 400, 0]).astype(np.float)
        self.B_birth[...,3] = np.diag([10,10,10,10]).astype(np.float)
        self.P_birth[...,3] = self.B_birth[...,3].dot(self.B_birth[...,3].T)

        self.lambda_c = 60
        self.range_c = np.array([[-1000,1000],
                                [-1000,1000]])
        self.pdf_c = 1/ np.prod(self.range_c[:,1] - self.range_c[:,0])

    def gen_truth(self):
        K = 100
        total_track = 12
        truth ={'K':K,
                'X':np.empty((self.x_dim,K,total_track))*np.nan,
                'N':np.zeros((K,),dtype=np.int8),
                'track_list':np.empty((K,total_track))*np.nan,
                'total_track':total_track
        }
        xstart =np.zeros((self.x_dim,total_track),dtype=np.float)
        xstart[:, 0] = np.array([0, 0, 0, -10])
        xstart[:, 1] = np.array([400, -10, -600, 5])
        xstart[:, 2] = np.array([-800, 20, -200, -5])
        xstart[:, 3] = np.array([400, -7, -600, -4])
        xstart[:, 4] = np.array([400, -2.5, -600, 10])
        xstart[:, 5] = np.array([0, 7.5, 0, -5])
        xstart[:, 6] = np.array([-800, 12, -200, 7])
        xstart[:, 7] = np.array([-200, 15, 800, -10])
        xstart[:, 8] = np.array([-800, 3, -200, 15])
        xstart[:, 9] = np.array([-200, -3, 800, -15])
        xstart[:, 10] = np.array([0, -20, 0, -15])
        xstart[:, 11] = np.array([-200, 15, 800, -5])
        # define birth and death time
        bd_time = np.array([[0,0,10,10,10,20,20,20,30,30,40,40],
                            [70,70,75,75,80,80,85,90,90,95,100,100]])
        for target_num in range(total_track):
            tstate = xstart[:,target_num]
            for k in range(bd_time[0,target_num],min(bd_time[1,target_num],K)):
                tstate =self.F.dot(tstate)+np.sqrt(self.Q).dot(np.random.randn(self.x_dim,))
                truth['X'][:,k,target_num] =tstate
                truth['track_list'][k,target_num]=target_num
                truth['N'][k] += 1
        return truth

    def gen_truth2(self):
        K = 100
        total_track = 1
        truth ={'K':K,
                'X':np.empty((self.x_dim,K,total_track))*np.nan,
                'N':np.zeros((K,),dtype=np.int8),
                'track_list':np.empty((K,total_track))*np.nan,
                'total_track':total_track
        }
        xstart =np.zeros((self.x_dim,total_track),dtype=np.float)
        xstart[:, 0] = np.array([400, 0, -400, -0])

        # define birth and death time
        bd_time = np.array([[0,],
                            [100]])
        for target_num in range(total_track):
            tstate = xstart[:,target_num]
            for k in range(bd_time[0,target_num],min(bd_time[1,target_num],K)):
                tstate =self.F.dot(tstate)+np.sqrt(self.Q).dot(np.random.randn(self.x_dim,))
                truth['X'][:,k,target_num] =tstate
                truth['track_list'][k,target_num]=target_num
                truth['N'][k] += 1
        return truth

    def gen_meas(self,truth):
        meas = {
            'K':truth['K'],
            'Z':[] #'Z': []  # np.empty((self.dim_obs, truth['K'], truth['total_tracks'])) * np.nan,
        }
        zero_mean = np.zeros((self.z_dim, ))
        for k in range(meas['K']):
            Z_k = None
            detected = np.random.rand(truth['N'][k],) <= self.P_D
            x = truth['X'][:,k,:]
            present_and_detected = ~ np.isnan(x.sum(axis=0)) # 互补符号， 1（第二位)的互补符号是-2,而0的互补符号是-1 但是在序列里至是给True/False取反
            #这里相当于 p_d = p_d & d,上一个语句只能说明state is present，下一个语句表明 stated is detected
            present_and_detected[present_and_detected == True] &= detected
            x = x[:,present_and_detected]
            # generated measurements and clutter
            r = np.random.multivariate_normal(zero_mean,self.R,size = x.shape[1]).T
            Z_k = self.H.dot(x)+r
            N_c = np.random.poisson(self.lambda_c)
            # [-1e3, 1e3;-1e3,1e3]*[-1;1]
            bounds = np.diag(self.range_c.dot(np.array([-1,1])))
            clutter = -1000.0+bounds.dot(np.random.rand(self.z_dim,N_c))
            # concatenate the sequence with column-wise
            Z_k = np.hstack((Z_k,clutter)) if Z_k is not None else clutter
            meas['Z'].append(Z_k)
        return meas