diff options
Diffstat (limited to 'System_Python')
| -rw-r--r-- | System_Python/system_swingup_test.py | 24 |
1 files changed, 12 insertions, 12 deletions
diff --git a/System_Python/system_swingup_test.py b/System_Python/system_swingup_test.py index 1b30c41..e4d3a72 100644 --- a/System_Python/system_swingup_test.py +++ b/System_Python/system_swingup_test.py @@ -82,8 +82,8 @@ class SwingUpEnv(): state = self.state
x, x_dot, theta, theta_dot = state
force = self.force_mag * action[0]
- self.sys.adjust(force)
-
+ self.sys.adjust(force)
+
costheta = math.cos(theta)
sintheta = math.sin(theta)
@@ -94,12 +94,12 @@ class SwingUpEnv(): else:
self.up_time = 0
- new_theta, new_x = self.sys.measure()
- new_theta = radians(new_theta)
- theta_dot = (new_theta - theta) / self.tau
+ new_theta, new_x = self.sys.measure()
+ new_theta = radians(new_theta)
+ theta_dot = (new_theta - theta) / self.tau
x_dot = (new_x - x) / self.tau
- self.state = (new_x, x_dot, new_theta, theta_dot)
-
+ self.state = (new_x, x_dot, new_theta, theta_dot)
+
done = x < -self.x_threshold \
or x > self.x_threshold \
or theta_dot < -self.theta_dot_threshold \
@@ -121,10 +121,10 @@ class SwingUpEnv(): return np.array(self.state), reward, done, {'max_up_time' : self.max_up_time}
def reset(self):
- self.sys.return_home()
- time.sleep(1)
- self.state = (0, 0, np.pi, 0)
-
+ self.sys.return_home()
+ time.sleep(1)
+ self.state = (0, 0, np.pi, 0)
+
self.up_time = 0
self.max_up_time = 0
self.up = False
@@ -219,7 +219,7 @@ class sarsaAgent: self.r_last = np.copy(r)
# This is the environment
-env = swingUp.SwingUpEnv()
+env = SwingUpEnv()
# For simplicity, we only consider forces of -1 and 1
numActions = 2
|