gaspersic/Gymnasium
1
0
Fork 0
mirror of https://github.com/Farama-Foundation/Gymnasium.git synced 2025-08-30 01:50:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

Updated Calculation Order (#1019)

Browse Source 
* Updated order Calculation

Updated order of `x` and `theta` calculations so that they are no longer one timestamp behind `x_dot` and `theta_dot`.

https://github.com/openai/gym/issues/1018

* Added semi-implicit euler option

* Got implicit and standard euler mixed up

* switched default option
This commit is contained in:
Alex
2018-09-21 17:19:40 -05:00
committed by pzhokhov
parent b2149e9041
commit 42f9e14c00
1 changed files with 11 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
gym/envs/classic_control/cartpole.py
View File

@@ -62,6 +62,7 @@ class CartPoleEnv(gym.Env):
self.polemass_length = (self.masspole * self.length) self.polemass_length = (self.masspole * self.length)
self.force_mag = 10.0 self.force_mag = 10.0
self.tau = 0.02 # seconds between state updates self.tau = 0.02 # seconds between state updates
self.kinematics_integrator = 'euler'
# Angle at which to fail the episode # Angle at which to fail the episode
self.theta_threshold_radians = 12 * 2 * math.pi / 360 self.theta_threshold_radians = 12 * 2 * math.pi / 360
@@ -97,10 +98,16 @@ class CartPoleEnv(gym.Env):
temp = (force + self.polemass_length * theta_dot * theta_dot * sintheta) / self.total_mass temp = (force + self.polemass_length * theta_dot * theta_dot * sintheta) / self.total_mass
thetaacc = (self.gravity * sintheta - costheta* temp) / (self.length * (4.0/3.0 - self.masspole * costheta * costheta / self.total_mass)) thetaacc = (self.gravity * sintheta - costheta* temp) / (self.length * (4.0/3.0 - self.masspole * costheta * costheta / self.total_mass))
xacc = temp - self.polemass_length * thetaacc * costheta / self.total_mass xacc = temp - self.polemass_length * thetaacc * costheta / self.total_mass
x = x + self.tau * x_dot if self.kinematics_integrator == 'euler':
x_dot = x_dot + self.tau * xacc x = x + self.tau * x_dot
theta = theta + self.tau * theta_dot x_dot = x_dot + self.tau * xacc
theta_dot = theta_dot + self.tau * thetaacc theta = theta + self.tau * theta_dot
theta_dot = theta_dot + self.tau * thetaacc
else: # semi-implicit euler
x_dot = x_dot + self.tau * xacc
x = x + self.tau * x_dot
theta_dot = theta_dot + self.tau * thetaacc
theta = theta + self.tau * theta_dot
self.state = (x,x_dot,theta,theta_dot) self.state = (x,x_dot,theta,theta_dot)
done = x < -self.x_threshold \ done = x < -self.x_threshold \
or x > self.x_threshold \ or x > self.x_threshold \