forked from knagara/SLAMwithCameraIMUforPython
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathsensor.py
205 lines (169 loc) · 6.3 KB
/
sensor.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
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
# -*- coding: utf-8 -*-
"""
sensor.py
author: Keita Nagara 永良慶太 (University of Tokyo) <nagara.keita()gmail.com>
This class is called from "Main.py", and process sensor data.
"""
from math import *
import numpy as np
import Util
import KF
class Sensor:
def __init__(self,_state):
#state.py
self.state = _state
self.init()
def init(self):
#variables
self.isFirstTime = True
self.time = 0.0
self.time1 = 0.0
self.accel = np.array([])
self.accel_g = np.array([])
self.gravity = np.array([]) #
self.magnet = np.array([])
self.magnet_fixed = np.array([])
self.gyro = np.array([])
self.gyro_diff = np.array([])
self.orientation = np.array([0.0,0.0,0.0])
self.orientation_g = np.array([0.0,0.0,0.0])
self.orientation_gyro = np.array([0.0,0.0,0.0])
self.rotX_ = np.identity(3)
self.rotY_ = np.identity(3)
self.rotX = np.identity(3)
self.rotY = np.identity(3)
self.rotZ = np.identity(3)
self.rot = np.identity(3)
self.I = np.identity(3)
self.P = np.array([0.0,0.0,0.0]) # covariance matrix of KF for orientation
self.Q = np.diag([0.1,0.1,0.1]) # noise of KF for orientation
self.R = np.diag([0.01,0.01,0.01]) # noise of KF for orientation
self.centrifugal = np.array([0.0,0.0,0.0]) #
self.tangential = np.array([0.0,0.0,0.0]) #
self.r = np.array([0.0,0.0,0.0])
self.v = np.array([0.0,0.0,0.0])
self.v1 = np.array([0.0,0.0,0.0])
#Set new data and Execute all functions
def processData(self,data):
self.time1 = self.time
self.time = (float(long(data[0]) / 1000.0))
self.accel = np.array([float(data[1]),float(data[2]),float(data[3])])
self.orientation_g = np.array([float(data[4]),float(data[5]),float(data[6])])
self.magnet = np.array([float(data[7]),float(data[8]),float(data[9])])
self.gyro = np.array([float(data[10]),float(data[11]),float(data[12])])
#self.gyro_diff = np.array([float(data[13]),float(data[14]),float(data[15])])
#self.calcOrientation()
#self.calcRotationMatrix()
#self.calcGlobalAcceleration()
self.pushDataToState()
if(self.isFirstTime):
self.isFirstTime = False
#Calc orientation
def calcOrientation(self):
self.calcOrientationByGravity()
if(self.isFirstTime):
self.orientation = self.orientation_g
else:
t = self.time - self.time1
matrixGyro2Euler = Util.matrixGyro2Euler(self.orientation[0],self.orientation[1]) * t
#Kalman Filter
resultKF = KF.execKF1(self.orientation_g, self.gyro, self.orientation, self.P, self.I, matrixGyro2Euler, self.I, self.Q, self.R)
self.orientation = resultKF[0]
self.P = resultKF[1]
if(self.orientation[0]>=pi):
self.orientation[0] -= pi*2.0
if(self.orientation[1]>=pi):
self.orientation[1] -= pi*2.0
if(self.orientation[2]>=pi):
self.orientation[2] -= pi*2.0
if(self.orientation[0]<-pi):
self.orientation[0] += pi*2.0
if(self.orientation[1]<-pi):
self.orientation[1] += pi*2.0
if(self.orientation[2]<-pi):
self.orientation[2] += pi*2.0
#Calc orientation by using gyro
def calcOrientationByGyro(self):
if(self.isFirstTime):
self.orientation_gyro = self.orientation_g
else:
t = self.time - self.time1
gyroEuler = np.dot(Util.matrixGyro2Euler(self.orientation_gyro[0],self.orientation_gyro[1]),self.gyro)
self.orientation_gyro = self.orientation_gyro + gyroEuler * t
if(self.orientation_gyro[0]>=pi):
self.orientation_gyro[0] -= pi*2.0
if(self.orientation_gyro[1]>=pi):
self.orientation_gyro[1] -= pi*2.0
if(self.orientation_gyro[2]>=pi):
self.orientation_gyro[2] -= pi*2.0
if(self.orientation_gyro[0]<-pi):
self.orientation_gyro[0] += pi*2.0
if(self.orientation_gyro[1]<-pi):
self.orientation_gyro[1] += pi*2.0
if(self.orientation_gyro[2]<-pi):
self.orientation_gyro[2] += pi*2.0
#Calc orientation by using gravity and magnet
#return orientation
#see also "Studies on Orientation Measurement in Sports Using Inertial and Magnetic Field Sensors"
# https://www.jstage.jst.go.jp/article/sposun/22/2/22_255/_pdf
def calcOrientationByGravity(self):
#x roll
self.orientation_g[0] = atan2(self.gravity[1],self.gravity[2])
#y pitch (-90 ～ +90)
self.orientation_g[1] = atan2(-self.gravity[0],hypot(self.gravity[1],self.gravity[2]))
#y pitch (-180 ～ +180)
"""
if(self.gravity[2]<0): #decided by z axis
self.orientation_g[1] = atan2(-self.gravity[0],-hypot(self.gravity[1],self.gravity[2]))
else:
self.orientation_g[1] = atan2(-self.gravity[0],hypot(self.gravity[1],self.gravity[2]))
"""
#z yaw
self.rotX_ = Util.rotationMatrixX(self.orientation_g[0])
self.rotY_ = Util.rotationMatrixY(self.orientation_g[1])
self.magnet_fixed = np.dot(np.dot(self.rotY_,self.rotX_),self.magnet)
self.orientation_g[2] = atan2(-self.magnet_fixed[1],self.magnet_fixed[0])
#Calc rotation matrix from orientation
def calcRotationMatrix(self):
#Rotation matrix R(Z)R(Y)R(X)
self.rotX = Util.rotationMatrixX(self.orientation[0])
self.rotY = Util.rotationMatrixY(self.orientation[1])
self.rotZ = Util.rotationMatrixZ(self.orientation[2])
self.rot = np.dot(self.rotZ,np.dot(self.rotY,self.rotX))
#Remove Centrifugal Accel
def removeCentrifugalAndTangentialAccel(self):
#Angular velocity
w = self.gyro
#Angular acceleration
#wa = self.gyro_diff
#wn2 (norm of gyro vector)^2
wn2 = pow(np.linalg.norm(w),2)
#norm of global v
vn = np.linalg.norm(self.state.v)
#centrifugal
centrifugal_x = np.array([0.0,0.0,0.0])
centrifugal_y = np.array([0.0,0.0,0.0])
if(w[0] > 0.3 or w[0] < -0.3):
#v of x
vx = np.array([0.0,vn*sin(self.orientation[0]),vn*cos(self.orientation[0])])
#w of x
wx = np.array([w[0],0.0,0.0])
#w*(w*r) of x
centrifugal_x = np.cross(wx,np.cross(wx,np.cross(vx,wx)/wn2))
if(w[1] > 0.3 or w[1] < -0.3):
#v of y
vy = np.array([-vn*sin(self.orientation[1]),0.0,vn*cos(self.orientation[1])])
#w of y
wy = np.array([0.0,w[1],0.0])
#w*(w*r) of y
centrifugal_y = np.cross(wy,np.cross(wy,np.cross(vy,wy)/wn2))
#a = a - w*(w*r)
self.centrifugal = centrifugal_x + centrifugal_y
self.accel = self.accel - self.centrifugal
#Calc accel in global coordinates by using orientation
def calcGlobalAcceleration(self):
#accel in global = R(Z)R(Y)R(X) * accel
self.accel_g = np.dot(self.rot,self.accel)
#Push all data to State class
def pushDataToState(self):
self.state.setSensorData(self.time, self.accel, self.orientation_g, self.gyro)