配置一个RL Agent

在之前的教程中，我们看到了如何使用 Stable-Baselines3 库训练RL智能体以解决cartpole平衡任务。在本教程中，我们将看到如何配置训练过程以使用不同的RL库和不同的训练算法。

在目录 scripts/reinforcement_learning 中，您将找到不同RL库的脚本。这些按照库名称命名的子目录进行组织。每个子目录包含该库的训练和回放脚本。

要为特定任务配置学习库，您需要为学习代理创建一个配置文件。这个配置文件用于创建学习代理的实例，并用于配置训练过程。类似于在:ref:`tutorial-register-rl-env-gym`教程中显示的环境注册，您可以使用``gymnasium.register``方法注册学习代理。

代码

作为示例，我们将查看``isaaclab_tasks``包中为任务``Isaac-Cartpole-v0``包含的配置。这是我们在:ref:`tutorial-run-rl-training`教程中使用的相同任务。

gym.register(
    id="Isaac-Cartpole-v0",
    entry_point="isaaclab.envs:ManagerBasedRLEnv",
    disable_env_checker=True,
    kwargs={
        "env_cfg_entry_point": f"{__name__}.cartpole_env_cfg:CartpoleEnvCfg",
        "rl_games_cfg_entry_point": f"{agents.__name__}:rl_games_ppo_cfg.yaml",
        "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpolePPORunnerCfg",
        "rsl_rl_with_symmetry_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpolePPORunnerWithSymmetryCfg",
        "skrl_cfg_entry_point": f"{agents.__name__}:skrl_ppo_cfg.yaml",
        "sb3_cfg_entry_point": f"{agents.__name__}:sb3_ppo_cfg.yaml",
    },

代码详解

在属性`kwargs`下，我们可以看到不同学习库的配置。 键是库的名称，值是配置实例的路径。 这个配置实例可以是字符串、类或类的实例。 例如，键`”rl_games_cfg_entry_point”的值是指向RL-Games库配置YAML文件的字符串。 同时，键”rsl_rl_cfg_entry_point”`的值指向RSL-RL库的配置类。

用于指定代理配置类的模式与用于指定环境配置入口点的模式非常相似。这意味着以下内容是等效的：

指定配置入口点为字符串

from . import agents

gym.register(
   id="Isaac-Cartpole-v0",
   entry_point="isaaclab.envs:ManagerBasedRLEnv",
   disable_env_checker=True,
   kwargs={
      "env_cfg_entry_point": f"{__name__}.cartpole_env_cfg:CartpoleEnvCfg",
      "rsl_rl_cfg_entry_point": f"{agents.__name__}.rsl_rl_ppo_cfg:CartpolePPORunnerCfg",
   },
)

指定配置入口点为一个类

from . import agents

gym.register(
   id="Isaac-Cartpole-v0",
   entry_point="isaaclab.envs:ManagerBasedRLEnv",
   disable_env_checker=True,
   kwargs={
      "env_cfg_entry_point": f"{__name__}.cartpole_env_cfg:CartpoleEnvCfg",
      "rsl_rl_cfg_entry_point": agents.rsl_rl_ppo_cfg.CartpolePPORunnerCfg,
   },
)

第一个代码块是指定配置入口的首选方式。第二个代码块等同于第一个代码块，但会导致配置类的导入，从而减慢导入时间。这就是为什么我们建议使用字符串作为配置入口的原因。

所有位于 scripts/reinforcement_learning 目录中的脚本都默认配置为从 kwargs 字典中读取 <library_name>_cfg_entry_point 以获取配置实例。

例如，以下代码块显示了 train.py 脚本如何读取 Stable-Baselines3 库的配置实例：

使用SB3的train.py代码

# Copyright (c) 2022-2025, The Isaac Lab Project Developers (https://github.com/isaac-sim/IsaacLab/blob/main/CONTRIBUTORS.md).
# All rights reserved.
#
# SPDX-License-Identifier: BSD-3-Clause


"""Script to train RL agent with Stable Baselines3."""

"""Launch Isaac Sim Simulator first."""

import argparse
import contextlib
import signal
import sys
from pathlib import Path

from isaaclab.app import AppLauncher

# add argparse arguments
parser = argparse.ArgumentParser(description="Train an RL agent with Stable-Baselines3.")
parser.add_argument("--video", action="store_true", default=False, help="Record videos during training.")
parser.add_argument("--video_length", type=int, default=200, help="Length of the recorded video (in steps).")
parser.add_argument("--video_interval", type=int, default=2000, help="Interval between video recordings (in steps).")
parser.add_argument("--num_envs", type=int, default=None, help="Number of environments to simulate.")
parser.add_argument("--task", type=str, default=None, help="Name of the task.")
parser.add_argument(
    "--agent", type=str, default="sb3_cfg_entry_point", help="Name of the RL agent configuration entry point."
)
parser.add_argument("--seed", type=int, default=None, help="Seed used for the environment")
parser.add_argument("--log_interval", type=int, default=100_000, help="Log data every n timesteps.")
parser.add_argument("--checkpoint", type=str, default=None, help="Continue the training from checkpoint.")
parser.add_argument("--max_iterations", type=int, default=None, help="RL Policy training iterations.")
parser.add_argument("--export_io_descriptors", action="store_true", default=False, help="Export IO descriptors.")
parser.add_argument(
    "--keep_all_info",
    action="store_true",
    default=False,
    help="Use a slower SB3 wrapper but keep all the extra training info.",
)
# append AppLauncher cli args
AppLauncher.add_app_launcher_args(parser)
# parse the arguments
args_cli, hydra_args = parser.parse_known_args()
# always enable cameras to record video
if args_cli.video:
    args_cli.enable_cameras = True

# clear out sys.argv for Hydra
sys.argv = [sys.argv[0]] + hydra_args

# launch omniverse app
app_launcher = AppLauncher(args_cli)
simulation_app = app_launcher.app


def cleanup_pbar(*args):
    """
    A small helper to stop training and
    cleanup progress bar properly on ctrl+c
    """
    import gc

    tqdm_objects = [obj for obj in gc.get_objects() if "tqdm" in type(obj).__name__]
    for tqdm_object in tqdm_objects:
        if "tqdm_rich" in type(tqdm_object).__name__:
            tqdm_object.close()
    raise KeyboardInterrupt


# disable KeyboardInterrupt override
signal.signal(signal.SIGINT, cleanup_pbar)

"""Rest everything follows."""

import gymnasium as gym
import numpy as np
import os
import random
from datetime import datetime

import omni
from stable_baselines3 import PPO
from stable_baselines3.common.callbacks import CheckpointCallback, LogEveryNTimesteps
from stable_baselines3.common.vec_env import VecNormalize

from isaaclab.envs import (
    DirectMARLEnv,
    DirectMARLEnvCfg,
    DirectRLEnvCfg,
    ManagerBasedRLEnvCfg,
    multi_agent_to_single_agent,
)
from isaaclab.utils.dict import print_dict
from isaaclab.utils.io import dump_pickle, dump_yaml

from isaaclab_rl.sb3 import Sb3VecEnvWrapper, process_sb3_cfg

import isaaclab_tasks  # noqa: F401
from isaaclab_tasks.utils.hydra import hydra_task_config

# PLACEHOLDER: Extension template (do not remove this comment)


@hydra_task_config(args_cli.task, args_cli.agent)
def main(env_cfg: ManagerBasedRLEnvCfg | DirectRLEnvCfg | DirectMARLEnvCfg, agent_cfg: dict):
    """Train with stable-baselines agent."""
    # randomly sample a seed if seed = -1
    if args_cli.seed == -1:
        args_cli.seed = random.randint(0, 10000)

    # override configurations with non-hydra CLI arguments
    env_cfg.scene.num_envs = args_cli.num_envs if args_cli.num_envs is not None else env_cfg.scene.num_envs
    agent_cfg["seed"] = args_cli.seed if args_cli.seed is not None else agent_cfg["seed"]
    # max iterations for training
    if args_cli.max_iterations is not None:
        agent_cfg["n_timesteps"] = args_cli.max_iterations * agent_cfg["n_steps"] * env_cfg.scene.num_envs

    # set the environment seed
    # note: certain randomizations occur in the environment initialization so we set the seed here
    env_cfg.seed = agent_cfg["seed"]
    env_cfg.sim.device = args_cli.device if args_cli.device is not None else env_cfg.sim.device

    # directory for logging into
    run_info = datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
    log_root_path = os.path.abspath(os.path.join("logs", "sb3", args_cli.task))
    print(f"[INFO] Logging experiment in directory: {log_root_path}")
    # The Ray Tune workflow extracts experiment name using the logging line below, hence, do not change it (see PR #2346, comment-2819298849)
    print(f"Exact experiment name requested from command line: {run_info}")
    log_dir = os.path.join(log_root_path, run_info)
    # dump the configuration into log-directory
    dump_yaml(os.path.join(log_dir, "params", "env.yaml"), env_cfg)
    dump_yaml(os.path.join(log_dir, "params", "agent.yaml"), agent_cfg)
    dump_pickle(os.path.join(log_dir, "params", "env.pkl"), env_cfg)
    dump_pickle(os.path.join(log_dir, "params", "agent.pkl"), agent_cfg)

    # save command used to run the script
    command = " ".join(sys.orig_argv)
    (Path(log_dir) / "command.txt").write_text(command)

    # post-process agent configuration
    agent_cfg = process_sb3_cfg(agent_cfg, env_cfg.scene.num_envs)
    # read configurations about the agent-training
    policy_arch = agent_cfg.pop("policy")
    n_timesteps = agent_cfg.pop("n_timesteps")

    # set the IO descriptors output directory if requested
    if isinstance(env_cfg, ManagerBasedRLEnvCfg):
        env_cfg.export_io_descriptors = args_cli.export_io_descriptors
        env_cfg.io_descriptors_output_dir = log_dir
    else:
        omni.log.warn(
            "IO descriptors are only supported for manager based RL environments. No IO descriptors will be exported."
        )

    # create isaac environment
    env = gym.make(args_cli.task, cfg=env_cfg, render_mode="rgb_array" if args_cli.video else None)

    # convert to single-agent instance if required by the RL algorithm
    if isinstance(env.unwrapped, DirectMARLEnv):
        env = multi_agent_to_single_agent(env)

    # wrap for video recording
    if args_cli.video:
        video_kwargs = {
            "video_folder": os.path.join(log_dir, "videos", "train"),
            "step_trigger": lambda step: step % args_cli.video_interval == 0,
            "video_length": args_cli.video_length,
            "disable_logger": True,
        }
        print("[INFO] Recording videos during training.")
        print_dict(video_kwargs, nesting=4)
        env = gym.wrappers.RecordVideo(env, **video_kwargs)

    # wrap around environment for stable baselines
    env = Sb3VecEnvWrapper(env, fast_variant=not args_cli.keep_all_info)

    norm_keys = {"normalize_input", "normalize_value", "clip_obs"}
    norm_args = {}
    for key in norm_keys:
        if key in agent_cfg:
            norm_args[key] = agent_cfg.pop(key)

    if norm_args and norm_args.get("normalize_input"):
        print(f"Normalizing input, {norm_args=}")
        env = VecNormalize(
            env,
            training=True,
            norm_obs=norm_args["normalize_input"],
            norm_reward=norm_args.get("normalize_value", False),
            clip_obs=norm_args.get("clip_obs", 100.0),
            gamma=agent_cfg["gamma"],
            clip_reward=np.inf,
        )

    # create agent from stable baselines
    agent = PPO(policy_arch, env, verbose=1, tensorboard_log=log_dir, **agent_cfg)
    if args_cli.checkpoint is not None:
        agent = agent.load(args_cli.checkpoint, env, print_system_info=True)

    # callbacks for agent
    checkpoint_callback = CheckpointCallback(save_freq=1000, save_path=log_dir, name_prefix="model", verbose=2)
    callbacks = [checkpoint_callback, LogEveryNTimesteps(n_steps=args_cli.log_interval)]

    # train the agent
    with contextlib.suppress(KeyboardInterrupt):
        agent.learn(
            total_timesteps=n_timesteps,
            callback=callbacks,
            progress_bar=True,
            log_interval=None,
        )
    # save the final model
    agent.save(os.path.join(log_dir, "model"))
    print("Saving to:")
    print(os.path.join(log_dir, "model.zip"))

    if isinstance(env, VecNormalize):
        print("Saving normalization")
        env.save(os.path.join(log_dir, "model_vecnormalize.pkl"))

    # close the simulator
    env.close()


if __name__ == "__main__":
    # run the main function
    main()
    # close sim app
    simulation_app.close()

参数 --agent 用于指定要使用的学习库。这用于从 kwargs 字典中检索配置实例。您可以通过传递 --agent 参数来手动指定替代配置实例。

代码执行

由于cartpole平衡任务，RSL-RL 库提供两个配置实例，我们可以使用 --agent 参数来指定要使用的配置实例。

• 使用标准PPO配置进行训练：

  # standard PPO training
  ./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task Isaac-Cartpole-v0 --headless \
    --run_name ppo
  

• 使用具有对称增强的PPO配置进行训练：

  # PPO training with symmetry augmentation
  ./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task Isaac-Cartpole-v0 --headless \
    --agent rsl_rl_with_symmetry_cfg_entry_point \
    --run_name ppo_with_symmetry_data_augmentation
  
  # you can use hydra to disable symmetry augmentation but enable mirror loss computation
  ./isaaclab.sh -p scripts/reinforcement_learning/rsl_rl/train.py --task Isaac-Cartpole-v0 --headless \
    --agent rsl_rl_with_symmetry_cfg_entry_point \
    --run_name ppo_without_symmetry_data_augmentation \
    agent.algorithm.symmetry_cfg.use_data_augmentation=false
  

--run_name 参数用于指定运行的名称。这用于在``logs/rsl_rl/cartpole``目录中为运行创建一个目录。