-
Notifications
You must be signed in to change notification settings - Fork 1
/
MassSpringDamper_env_3v.py
190 lines (152 loc) · 6.79 KB
/
MassSpringDamper_env_3v.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
import gym
#import gym_environments.envs
from gym import error, spaces, utils
from gym.utils import seeding
import numpy as np
#UNCOMMENT if you intend to use the manipulator or GYM-ROS interface.
#import ros_message_listener.eavesdrop as eavesdrop
#from ManipulatorAction import ManipulatorAction
class MassSpringDamperEnv(gym.Env):
metadata = {
'render.modes': ['human', 'rgb_array'],
'video.frames_per_second' : 50
}
def __init__(self,cont_actions_bool = True, reward_func = None): #, spring_stiffness, damper_factor, mass, max_force4
if cont_actions_bool == False:
logger.warn("You are calling cont_actions_bool == False(discrete actions), but this functionality is not finished. -- any further steps are undefined behavior.")
self.x_treshold = 4.8
self.spring_stiffness = 0.25 #spring_stiffness
self.damper_factor = 0.25 #damper_factor
self.mass = 0.25 #mass
self.max_force = np.array([1]) #max_force
self.step_length = 0.1 # in seconds
obs_high = np.array([self.x_treshold,np.finfo(np.float32).max,np.finfo(np.float32).max])#,self.x_treshold,np.finfo(np.float32).max])
self.observation_space = spaces.Box(-obs_high,obs_high, dtype=np.float32)
self.action_space = spaces.Box(-self.max_force,self.max_force,dtype=np.float32)
self.state = None # Position x, velocity x_dot, distance to goal point error
self.goal_state = None # Some position with no velocity
if reward_func == None:
def r_f(x,x_dot,error):
if self.is_done():
return 10, True
else:
if self.is_terminated():
return -100, True
else:
return -np.abs(error)**2 -1, False
self.reward_func = r_f
else:
self.reward_func = reward_func
self.viewer = None
self.steps_taken = None
self.steps_beyond_done = None
def is_done(self):
state = self.state
x, x_dot, error = state
if (np.abs(error) < 0.01) and(np.abs(x_dot)< 0.001):
return True
else:
return False
def is_terminated(self):
state = self.state
x, x_dot, error = state
if x < -self.x_treshold or x > self.x_treshold:
return True
else:
return False
def step(self, action):
if not self.action_space.contains(action):
action = np.clip(action,-1,1)
assert self.action_space.contains(action), "%r (%s) invalid"%(action, type(action))
state = self.state
x, x_dot, error = state
acceleration = (-self.spring_stiffness*x -self.damper_factor*x_dot + action)/self.mass
x_dot += acceleration*self.step_length
x += x_dot*self.step_length
x_goal,x_dot_goal = self.goal_state
error = x_goal-x[0]
self.state = (x[0],x_dot[0],error)
done = False
reward, done = self.reward_func(x,x_dot,error)
if done:
if self.steps_beyond_done is None:
self.steps_beyond_done = 0
else:
if self.steps_beyond_done == 0:
logger.warn("You are calling 'step()' even though this environment has already returned done = True. You should always call 'reset()' once you receive 'done = True' -- any further steps are undefined behavior.")
self.steps_beyond_done += 1
reward = 0.0
if self.steps_taken == None:
self.steps_taken = 1
else:
self.steps_taken += 1
if self.steps_taken == 200:
done = True
return np.array(self.state), reward, done, {}
def reset(self):
goal_state = [np.random.uniform(low = -self.x_treshold/1.2, high = self.x_treshold/1.2),0.00]
self.goal_state =(goal_state[0], goal_state[1])
start_point = np.random.uniform(low = -self.x_treshold/1.5, high = self.x_treshold/1.5)
self.state = (start_point,0.0,goal_state[0]-start_point)
while (np.abs(self.state[2])<0.5):
start_point = np.random.uniform(low = -self.x_treshold/1.5, high = self.x_treshold/1.5)
self.state = (start_point,0.0,goal_state[0]-start_point)
self.steps_taken = None
self.steps_beyond_done = None
return np.array(self.state)
def render(self, mode='human'):
screen_width = 600
screen_height = 400
world_width = self.x_treshold*2
scale = screen_width/world_width
mass_y =100.0
mass_width = 50.0
mass_height = 30.0
if self.viewer is None:
from gym.envs.classic_control import rendering
self.viewer = rendering.Viewer(screen_width, screen_height)
l,r,t,b = -mass_width/2, mass_width/2,mass_height/2, -mass_height/2
axleoffset = mass_height/4.0
goal_mass = rendering.FilledPolygon([(l,b),(l,t),(r,t),(r,b)])
goal_mass.set_color(100,0,0)
self.goal_mass_trans = rendering.Transform()
goal_mass.add_attr(self.goal_mass_trans)
self.viewer.add_geom(goal_mass)
mass = rendering.FilledPolygon([(l,b),(l,t),(r,t),(r,b)])
self.mass_trans = rendering.Transform()
mass.add_attr(self.mass_trans)
self.viewer.add_geom(mass)
self.track = rendering.Line((0,mass_y), (screen_width,mass_y))
self.track.set_color(0,0,0)
self.viewer.add_geom(self.track)
### ADD SPRING
#p1 = [l,(t-b)/2.0]
#x = self.state
#mass_x = x[0]*scale+screen_width/2.0
#left = mass_x - mass_width/2.0
#y_m = mass_y + 0.25*mass_height
#y_t = mass_y + 0.5*mass_height
#y_b = mass_y
#self.p1 = rendering.Line(((3/4)*left,y_m),(left,y_m))
#self.p1.set_color(100,0,0)
#self.viewer.add_geom(self.p1)
if self.state is None: return None
x = self.state
#print(x, self.goal_state)
mass_x = x[0]*scale+screen_width/2.0 # MIDDLE OF MASS
self.mass_trans.set_translation(mass_x,mass_y)
self.goal_mass_trans.set_translation(self.goal_state[0]*scale + screen_width/2.0,mass_y)
return self.viewer.render(return_rgb_array = mode =='rgb_array')
def close(self):
if self.viewer:
self.viewer.close()
self.viewer = None
def get_obs(self):
return self.state
if __name__ == '__main__':
#Environment unit test!
#Cant be executed before environment is registered with gym, in envs/__init__.py
env = gym.make('MassSpringDamper-v0')
env.step(action_space.get_random_action())
env.reset()
++++2