Merge pull request #196 from maxspahn/ft-gymnasium-migration

Ft gymnasium migration

examples/albert.py

@@ -1,5 +1,5 @@
 import warnings
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.robots.generic_urdf.generic_diff_drive_robot import GenericDiffDriveRobot
 from urdfenvs.urdf_common.urdf_env import UrdfEnv
@@ -32,7 +32,7 @@ def run_albert(n_steps=1000, render=False, goal=True, obstacles=True):
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/boxer.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.scene_examples.obstacles import wall_obstacles
 import numpy as np
 
@@ -30,7 +30,7 @@ def run_boxer(n_steps=1000, render=False, goal=True, obstacles=True):
     print("Starting episode")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         print(ob['robot_0']['joint_state']['velocity'][0:2])
         history.append(ob)
     env.close()

examples/dual_arm.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 
@@ -16,7 +16,7 @@ def run_dual_arm(n_steps=5000, render=False, goal=True, obstacles=True):
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/generic_holonomic.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 import numpy as np
 import os
@@ -35,7 +35,7 @@ def run_generic_holonomic(n_steps=1000, render=False, goal=True, obstacles=True)
     print("Starting episode")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/heijn.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.scene_examples.obstacles import *
 from urdfenvs.scene_examples.goal import *
@@ -28,7 +28,7 @@ def run_point_robot(n_steps=1000, render=False, goal=True, obstacles=True):
         env.add_goal(splineGoal)
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/iris.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.robots.iris import IrisDrone
 
@@ -22,7 +22,7 @@ def run_iris(n_steps=3000, render=False, goal=True, obstacles=True):
             action += np.array([0, 1, 0, 1]) * 1
         if i > 248 and i < 298:
             action += np.array([0, 1, 0, 1]) * -1
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/jackal.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 
 from urdfenvs.scene_examples.obstacles import wall_obstacles
@@ -33,7 +33,7 @@ def run_jackal(n_steps=1000, render=False, goal=True, obstacles=True):
     print("Starting episode")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/keyboard_input.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.tiago import TiagoRobot
 
 from multiprocessing import Process, Pipe

examples/mobile_reacher.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 import numpy as np
 
@@ -25,7 +25,7 @@ def run_mobile_reacher(n_steps=1000, render=False, goal=True, obstacles=True):
         else:
             action[-1] = 0.01
             action[-2] = 0.01
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/multi_robot.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 from urdfenvs.robots.generic_urdf import GenericDiffDriveRobot
@@ -78,7 +78,7 @@ def run_multi_robot(n_steps=1000, render=False, obstacles=False, goal=False):
     print("Starting episode")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/n_link_urdf_reacher.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 import numpy as np
 
@@ -17,7 +17,7 @@ def run_n_link_reacher(n_steps=1000, render=False, goal=True, obstacles=True):
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/panda_reacher.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 from urdfenvs.scene_examples.goal import dynamicGoal
@@ -46,7 +46,7 @@ def run_panda(n_steps=1000, render=False, goal=True, obstacles=True):
             action[7] = 0.02
             action[8] = 0.02
         collision_links_position: dict = env.collision_links_poses(position_only=True)
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/point_robot.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 import numpy as np
 
@@ -32,7 +32,7 @@ def run_point_robot(n_steps=1000, render=False, goal=True, obstacles=True):
     history = []
     env.reconfigure_camera(2.0, 0.0, -90.01, (0, 0, 0))
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history

examples/point_robot_3dlidar_sensor.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.scene_examples.obstacles import (
     sphereObst1,
@@ -46,7 +46,7 @@ def run_point_robot_with_lidar(
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         # Access the lidar observation
         #_ = ob["robot_0"]["LidarSensor"]
 

examples/point_robot_free_space_decomp.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import pprint
 import numpy as np
 from urdfenvs.scene_examples.obstacles import (
@@ -41,7 +41,7 @@ def run_point_robot_with_freespacedecomp(
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         #print(ob['robot_0']['FreeSpaceDecompSensor'])
         # Access the lidar observation
         #_ = ob["robot_0"]["LidarSensor"]

examples/point_robot_full_sensor.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.urdf_common.urdf_env import UrdfEnv
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 from urdfenvs.sensors.full_sensor import FullSensor

examples/point_robot_lidar_sensor.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 from urdfenvs.scene_examples.obstacles import (
     sphereObst1,
@@ -40,7 +40,7 @@ def run_point_robot_with_lidar(
     print(f"Initial observation : {ob}")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         # Access the lidar observation
         #_ = ob["robot_0"]["LidarSensor"]
 

examples/point_robot_obstacle_sensor.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf import GenericUrdfReacher
 from urdfenvs.sensors.obstacle_sensor import ObstacleSensor
 from urdfenvs.scene_examples.obstacles import (
@@ -36,7 +36,7 @@ def run_point_robot_with_obstacle_sensor(n_steps=1000, render=False, obstacles=T
     history = []
     for _ in range(n_steps):
         action = defaultAction
-        ob, reward, done, info = env.step(action)
+        ob, *_ = env.step(action)
         # In observations, information about obstacles is stored in ob['obstacleSensor']
         history.append(ob)
     env.close()

examples/point_robot_reward.py

@@ -10,7 +10,7 @@
 
 
 class InverseDistanceDenseReward(Reward):
-    def calculateReward(self, observation: dict) -> float:
+    def calculate_reward(self, observation: dict) -> float:
         goal = observation['robot_0']['FullSensor']['goals'][1]['position']
         position = observation['robot_0']['joint_state']['position']
         return 1.0/np.linalg.norm(goal-position)

examples/point_robot_sdf_sensor.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 import numpy as np
 
 from urdfenvs.urdf_common.urdf_env import UrdfEnv

examples/prius.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.prius import Prius
 import numpy as np
 
@@ -17,7 +17,7 @@ def run_prius(n_steps=1000, render=False, goal=True, obstacles=True):
     print(f"Initial observation : {ob}")
     history = []
     for i in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         if ob['robot_0']['joint_state']['steering'] > 0.2:
             action[1] = 0
         history.append(ob)

examples/tiago.py

@@ -1,4 +1,4 @@
-import gym
+import gymnasium as gym
 from urdfenvs.robots.generic_urdf.generic_diff_drive_robot import GenericDiffDriveRobot
 import numpy as np
 
@@ -48,7 +48,7 @@ def run_tiago(n_steps=1000, render=False, goal=True, obstacles=True):
     action[14] = 0.0 # right arm shoulder
     action[22] = 0.01 # finger joint
     vel0 = np.zeros(env.n())
-    ob = env.reset()
+    ob, *_ = env.reset()
     print("base: ", ob['robot_0']["joint_state"]["position"][0:3])
     print("torso: ", ob['robot_0']["joint_state"]["position"][3])
     print("head: ", ob['robot_0']["joint_state"]["position"][4:6])
@@ -57,7 +57,7 @@ def run_tiago(n_steps=1000, render=False, goal=True, obstacles=True):
     print("Starting episode")
     history = []
     for _ in range(n_steps):
-        ob, _, _, _ = env.step(action)
+        ob, *_ = env.step(action)
         history.append(ob)
     env.close()
     return history