Hugging Face 推出 Microduck 开源机器人：支持强化学习与低成本具身智能探索

pip install lerobot requests opencv-python

import cv2
import json
import requests
import time
import base64

# 配置您的 n1n.ai API 凭证
N1N_API_KEY = "your_n1n_api_key_here"
N1N_API_URL = "https://api.n1n.ai/v1/chat/completions"

def capture_robot_view():
    # 初始化摄像头（默认 0 为 Microduck 车载摄像头）
    cap = cv2.VideoCapture(0)
    ret, frame = cap.read()
    cap.release()
    if not ret:
        raise RuntimeError("无法获取 Microduck 摄像头画面")

    # 将图像编码为 base64 格式
    _, buffer = cv2.imencode('.jpg', frame)
    return base64.b64encode(buffer).decode('utf-8')

def get_next_action(base64_image, user_instruction):
    headers = {
        "Authorization": f"Bearer {N1N_API_KEY}",
        "Content-Type": "application/json"
    }

    payload = {
        "model": "claude-3-5-sonnet",
        "messages": [
            {
                "role": "system",
                "content": (
                    "你扮演 Microduck 机器人的高层大脑。"
                    "请分析摄像头图像并决定当前应执行的低层动作策略。"
                    "请仅返回一个包含 'policy_name' 和 'duration' 的 JSON 对象。"
                    "可选策略列表: ['locate_target', 'approach_object', 'nudge_forward', 'stop']。"
                )
            },
            {
                "role": "user",
                "content": [
                    {
                        "type": "text",
                        "text": f"任务指令: {user_instruction}"
                    },
                    {
                        "type": "image_url",
                        "image_url": {
                            "url": f"data:image/jpeg;base64,{base64_image}"
                        }
                    }
                ]
            }
        ],
        "response_format": { "type": "json_object" }
    }

    response = requests.post(N1N_API_URL, headers=headers, json=payload)
    response.raise_for_status()
    result = response.json()

    content = result['choices'][0]['message']['content']
    return json.loads(content)

def execute_policy(policy_name, duration):
    print(f"[Microduck] 正在执行策略: {policy_name}，持续时间 {duration} 秒...")
    # 在实际硬件中，此处将通过 LeRobot 加载对应的 PyTorch 权重
    # 示例: policy = lerobot.load_policy(policy_name)
    # policy.step()
    time.sleep(duration)
    print("[Microduck] 策略执行完毕。")

if __name__ == "__main__":
    instruction = "把黄色玩具推开。"

    try:
        print("[Microduck] 正在捕获当前环境画面...")
        img_data = capture_robot_view()

        print("[Microduck] 正在发送画面至 n1n.ai API 进行决策...")
        decision = get_next_action(img_data, instruction)

        print(f"[Microduck] 收到决策指令: {decision}")
        execute_policy(decision['policy_name'], decision['duration'])

    except Exception as e:
        print(f"运行出错: {e}")