Adds render option for acceleration environment for pointRobot and pl…

…anarArm robot

examples/nLink_reacher.py

@@ -6,8 +6,8 @@
 def main():
     n = 3
     # env = gym.make('nLink-reacher-acc-v0', n=n, dt=0.01)
-    env = gym.make('nLink-reacher-tor-v0', n=n, dt=0.01, k=2.1)
-    defaultAction = np.ones(n) * 8
+    env = gym.make('nLink-reacher-acc-v0', render=True, n=n, dt=0.01)
+    defaultAction = np.ones(n) * 8 * 0.01
     n_episodes = 1
     n_steps = 1000
     cumReward = 0.0
@@ -17,7 +17,6 @@ def main():
         for i in range(n_steps):
             action = env.action_space.sample()
             action = defaultAction
-            env.render()
             ob, reward, done, info = env.step(action)
             cumReward += reward
 

examples/point_robot.py

@@ -4,7 +4,7 @@
 
 
 def main():
-    env = gym.make('point-robot-acc-v0', dt=0.01)
+    env = gym.make('point-robot-acc-v0', render=True, dt=0.01)
     defaultAction = [0.0, 0.10]
     initPos = np.array([1.0, 0.0])
     initVel = np.array([-1.0, 0.0])
@@ -17,7 +17,6 @@ def main():
         for i in range(n_steps):
             action = env.action_space.sample()
             action = defaultAction
-            env.render()
             ob, reward, done, info = env.step(action)
             cumReward += reward
             if done:

nLinkReacher/envs/acc.py

@@ -18,7 +18,7 @@ class NLinkAccReacherEnv(core.Env):
     MAX_POS = pi
     MAX_ACC = 9 * pi
 
-    def __init__(self, n=2, dt=0.01):
+    def __init__(self, render=False, n=2, dt=0.01):
         self._n = n
         self.viewer = None
         limUpPos = [self.MAX_POS for i in range(n)]
@@ -33,6 +33,7 @@ def __init__(self, n=2, dt=0.01):
         self.state = None
         self.seed()
         self._dt = dt
+        self._render = render
 
     def dt(self):
         return self._dt
@@ -59,6 +60,8 @@ def step(self, a):
         self.state = ns
         terminal = self._terminal()
         reward = -1.0 if not terminal else 0.0
+        if self._render:
+            self.render()
         return (self._get_ob(), reward, terminal, {})
 
     def _get_ob(self):

pointRobot/envs/acc.py

@@ -16,7 +16,7 @@ class PointRobotAccEnv(core.Env):
     MAX_POS = 10
     MAX_ACC = 10
 
-    def __init__(self, n=2, dt=0.01):
+    def __init__(self, render=False, n=2, dt=0.01):
         self._n = n
         self.viewer = None
         limUpPos = [self.MAX_POS for i in range(n)]
@@ -31,6 +31,7 @@ def __init__(self, n=2, dt=0.01):
         self.state = None
         self.seed()
         self._dt = dt
+        self._render = render
 
     def dt(self):
         return self._dt
@@ -50,6 +51,8 @@ def step(self, a):
         self.state = ns
         terminal = self._terminal()
         reward = -1.0 if not terminal else 0.0
+        if self._render:
+            self.render()
         return (self._get_ob(), reward, terminal, {})
 
     def _get_ob(self):