Robust Robot Manipulation Tasks

Robust Robot Manipulation Tasks#

Robust Robot Manipulation Tasks#

TasksRobust type

Robust State

Robust Action

Robust Reward

RobustLift

✅

✅

✅

RobustDoor

✅

✅

✅

RobustNutAssembly

✅

✅

✅

RobustPickPlace

✅

✅

✅

RobustStack

✅

✅

✅

RobustWipe

✅

✅

✅

RobustToolHang

✅

✅

✅

RobustTwoArmLift

✅

✅

✅

RobustTwoArmPegInHole

✅

✅

✅

RobustTwoArmHandover

✅

✅

✅

RobustTwoArmTransport

✅

✅

✅

MultiRobustDoor

✅

✅

✅

A Simple Example

This demonstrates the various functionalities of each controller available within robust_gymnasium.envs.robosuite.

For a given controller, the script runs through each dimension and executes a perturbation test_value from its neutral (stationary) value for a certain amount of time steps_per_action, then returns to all neutral values for steps_per_rest before proceeding with the next action dimension.

Example of the testing sequence of actions over time for the Pos / Ori (OSC_POSE) controller (without a gripper):

Expected sequential behavior:

  • OSC_POSE: Gripper moves sequentially and linearly in x, y, z direction, then rotates in x, y, z axes relative to the global coordinate frame.

  • OSC_POSITION: Gripper moves sequentially in x, y, z direction relative to the global frame.

  • IK_POSE: Gripper moves in x, y, z direction, then rotates in x, y, z axes relative to the local robot end effector frame.

  • JOINT_POSITION: Robot joints move sequentially in a controlled fashion.

  • JOINT_VELOCITY: Robot joints move sequentially in a controlled fashion.

  • JOINT_TORQUE: Joint torque controller acts sluggishly, as 0 torque will not guarantee a stable robot if it has non-zero velocity.

Code:

import robust_gymnasium.envs.robosuite as suite
from robust_gymnasium.envs.robosuite.controllers import load_controller_config
from robust_gymnasium.envs.robosuite.robots import Bimanual
from robust_gymnasium.envs.robosuite.utils.input_utils import *
from robust_gymnasium.configs.robust_setting import get_config
args = get_config().parse_args()

if __name__ == "__main__":
    options = {}
    print("Welcome to robust_gymnasium.envs.robosuite v{}!".format(suite.__version__))
    options["env_name"] = "Door"  # Choose environment

    if "TwoArm" in options["env_name"]:
        options["env_configuration"] = choose_multi_arm_config()
        options["robots"] = "Baxter" if options["env_configuration"] == "bimanual" else [choose_robots(exclude_bimanual=True) for _ in range(2)]
    else:
        options["robots"] = 'IIWA'  # Choose robot

    joint_dim = 6 if options["robots"] == "UR5e" else 7
    controller_name = 'JOINT_VELOCITY'
    options["controller_configs"] = suite.load_controller_config(default_controller=controller_name)

    controller_settings = {
        "OSC_POSE": [6, 6, 0.1],
        "OSC_POSITION": [3, 3, 0.1],
        "IK_POSE": [6, 6, 0.01],
        "JOINT_POSITION": [joint_dim, joint_dim, 0.2],
        "JOINT_VELOCITY": [joint_dim, joint_dim, -0.1],
        "JOINT_TORQUE": [joint_dim, joint_dim, 0.25],
    }

    action_dim = controller_settings[controller_name][0]
    num_test_steps = controller_settings[controller_name][1]
    test_value = controller_settings[controller_name][2]

    steps_per_action = 750
    steps_per_rest = 750
    robust_setting = {"robust_type": "state", "robust_config": args}
    env = suite.make(
        **options,
        has_renderer=True,
        has_offscreen_renderer=False,
        ignore_done=True,
        use_camera_obs=False,
        horizon=(steps_per_action + steps_per_rest) * num_test_steps,
        control_freq=20,
    )
    env.reset()
    env.viewer.set_camera(camera_id=0)

    n = sum(int(robot.action_dim / (action_dim + robot.gripper.dof)) for robot in env.robots)
    neutral = np.zeros(action_dim + robot.gripper.dof)

    count = 0
    while count < num_test_steps:
        action = neutral.copy()
        for i in range(steps_per_action):
            if controller_name in {"IK_POSE", "OSC_POSE"} and count > 2:
                vec = np.zeros(3)
                vec[count - 3] = test_value
                action[3:6] = vec
            else:
                action[count] = test_value
            total_action = np.tile(action, n)
            robust_input = {"action": total_action, "robust_type": "state", "robust_config": args}
            env.step(robust_input)
            env.render()
        for i in range(steps_per_rest):
            total_action = np.tile(neutral, n)
            robust_input = {"action": total_action, "robust_type": "state", "robust_config": args}
            env.step(robust_input)
            env.render()
        count += 1

    env.close()