-
Notifications
You must be signed in to change notification settings - Fork 11
/
bee_simulator.py
91 lines (64 loc) · 2.69 KB
/
bee_simulator.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
import numpy as np
import scipy as sp
n_sensors = 20
directions = np.linspace(-np.pi, np.pi, n_sensors, endpoint=False)
D = np.column_stack((np.cos(directions), np.sin(directions),
np.zeros(n_sensors)))
def image_motion_flow(T, R, D):
"""Calculate optic flow based on movement."""
P = T - (D.T * np.dot(D, T)).T
P -= np.cross(R, D)
return P
def rotary_flow(D, A):
"""Counterclockwise rotation."""
return -np.cross(D, A)
def translatory_flow(D, A):
return np.cross(np.cross(D, A), D)
def linear_range_model(U, P, w=1.0, n=0.0):
"""Eq 5 in Franz & Krapp"""
return np.sum(w * (np.sum(U * P, axis=1) + n))
def tn_axes(heading):
return np.array([[np.sin(heading - np.pi / 4.0),
np.cos(heading - np.pi / 4.0), 0],
[np.sin(heading + np.pi / 4.0),
np.cos(heading + np.pi / 4.0), 0]])
def get_flow2(heading, velocity):
"""This is the longwinded version that does all the flow calculations,
piece by piece. It can be refactored down to flow2() so use that for
performance benefit."""
T = np.array([velocity[0], velocity[1], 0.0]) # We are keeping
R = np.array([0.0, 0.0, 0.0])
P = image_motion_flow(T, R, D)
a = tn_axes(heading)
U_TN_1 = translatory_flow(D, a[0])
U_TN_2 = translatory_flow(D, a[1])
lr_1 = linear_range_model(U_TN_1, P, w=1.0/10.0)
lr_2 = linear_range_model(U_TN_2, P, w=1.0/10.0)
return np.array([lr_1, lr_2])
def get_flow(heading, velocity, pref_angle=np.pi/4):
A = np.array([[np.sin(heading - pref_angle),
np.cos(heading - pref_angle)],
[np.sin(heading + pref_angle),
np.cos(heading + pref_angle)]])
return np.dot(A, velocity)
def rotate(theta, r):
"""Return new heading after a rotation around Z axis."""
return (theta + r + np.pi) % (2.0 * np.pi) - np.pi
def thrust(theta, acceleration):
"""Thrust vector from current heading and acceleration
theta: clockwise radians around z-axis, where 0 is forward
acceleration: float where max speed is ....?!?
"""
return np.array([np.sin(theta), np.cos(theta)]) * acceleration
def get_next_state(heading, velocity, rotation, acceleration, drag=0.5):
"""Get new heading and velocity, based on relative rotation and
acceleration and linear drag."""
theta = rotate(heading, rotation)
v = velocity + thrust(theta, acceleration)
v -= drag * v
return theta, v
def run_simulation(N_outbound=500, N_inbound=1000):
N = N_outbound + N_inbound
for i in range(1, N-1):
headings[i], v[i, :], flow[i, :] = get_next_state(
headings[i-1], v[i-1, :], r[i], a, drag=0.25)