跳转到内容
3we

第一个机器人

30 秒 Hello World

Section titled “30 秒 Hello World”
import asyncio
from threewe import Robot


async def main():
    async with Robot(backend="mock") as robot:
        image = robot.get_image()
        await robot.move_to(x=2.0, y=1.0)
        pose = robot.get_pose()
        print(f"Arrived at: ({pose.x:.2f}, {pose.y:.2f})")


asyncio.run(main())

无需 ROS2，无需 Gazebo，无需 GPU。mock 后端在本地模拟一切，除 numpy 外零依赖。

切换后端

Section titled “切换后端”

更改一个参数即可在不同后端上运行：

# Local simulation (no dependencies)
Robot(backend="mock")


# Full physics simulation (Gazebo Harmonic)
Robot(backend="gazebo")


# GPU-accelerated training (Isaac Sim)
Robot(backend="isaac_sim")


# Real hardware (ROS2 Jazzy + Pi5 + ESP32)
Robot(backend="real")

所有后端共享完全相同的 API——零代码改动。

使用 Gymnasium 进行 RL 训练

Section titled “使用 Gymnasium 进行 RL 训练”
import gymnasium as gym
import threewe.gym  # Registers environments


env = gym.make("3we/Navigation-v1", render_mode="rgb_array")
obs, info = env.reset()


for _ in range(1000):
    action = env.action_space.sample()
    obs, reward, terminated, truncated, info = env.step(action)
    if terminated or truncated:
        obs, info = env.reset()


env.close()

VLM 驱动的导航

Section titled “VLM 驱动的导航”
import asyncio
from threewe import Robot


async def main():
    async with Robot(backend="mock") as robot:
        result = await robot.execute_instruction("find the red bottle and stop near it")
        print(f"Success: {result.success}")
        print(f"Description: {result.description}")


asyncio.run(main())

需要 pip install threewe[ai] 和 OPENAI_API_KEY 环境变量。

模仿学习数据采集

Section titled “模仿学习数据采集”
import asyncio
from threewe import Robot
from threewe.data import TrajectoryRecorder


async def main():
    recorder = TrajectoryRecorder(output_dir="trajectories/")


    async with Robot(backend="mock") as robot:
        recorder.start(robot)
        await robot.move_to(x=3.0, y=2.0)
        await robot.move_forward(1.0)
        await robot.rotate(1.57)
        recorder.stop()


asyncio.run(main())

API 速查表

Section titled “API 速查表”
方法描述
robot.get_image()相机图像 (480, 640, 3) uint8
robot.get_camera_image()与 get_image() 相同
robot.get_lidar_scan()360 度 LiDAR 扫描
robot.get_pose()地图坐标系中的位置 (x, y, theta)
robot.get_velocity()当前机体坐标系速度
robot.get_imu()IMU 传感器数据
robot.get_observation()ML 模型的标准化字典
await robot.move_to(x, y)使用路径规划导航到目标
await robot.move_forward(d)直线前进 d 米
await robot.rotate(a)旋转 a 弧度
robot.set_velocity(vx, vy, omega)直接速度指令
robot.stop()停止所有运动
await robot.execute_instruction(text)VLM 驱动的任务执行
robot.execute_action(action)执行策略网络输出

下一步

Section titled “下一步”