هوش مصنوعی و رباتیک

<span class="token comment"># نصب Google GenAI SDK (توصیه شده برای رباتیک)</span>
pip <span class="token function">install</span> <span class="token parameter variable">-U</span> google-genai

<span class="token comment"># یا نصب OpenAI SDK برای فرمت سازگار با OpenAI</span>
pip <span class="token function">install</span> <span class="token parameter variable">-U</span> openai

<span class="token comment"># رمزگذاری تصویر به base64</span>
<span class="token assign-left variable">IMAGE_BASE64</span><span class="token operator">=</span><span class="token variable">$(base64 <span class="token parameter variable">-w</span> <span class="token number">0</span> workspace.jpg)</span>

<span class="token function">curl</span> <span class="token parameter variable">-X</span> POST <span class="token punctuation">\</span>
  <span class="token string">"https://api.avalai.ir/v1beta/models/gemini-robotics-er-1.5-preview:generateContent"</span> <span class="token punctuation">\</span>
  <span class="token parameter variable">-H</span> <span class="token string">"Authorization: Bearer <span class="token variable">$AVALAI_API_KEY</span>"</span> <span class="token punctuation">\</span>
  <span class="token parameter variable">-H</span> <span class="token string">"Content-Type: application/json"</span> <span class="token punctuation">\</span>
  <span class="token parameter variable">-d</span> <span class="token string">'{
    "contents": [{
      "parts": [
        {
          "inlineData": {
            "mimeType": "image/jpeg",
            "data": "'</span>"<span class="token variable">${IMAGE_BASE64}</span><span class="token string">"'"</span>
          <span class="token punctuation">}</span>
        <span class="token punctuation">}</span>,
        <span class="token punctuation">{</span>
          <span class="token string">"text"</span><span class="token builtin class-name">:</span> <span class="token string">"Point to no more than 10 items in the image. Return JSON: [{<span class="token entity" title="\"">\"</span>point<span class="token entity" title="\"">\"</span>: [y, x], <span class="token entity" title="\"">\"</span>label<span class="token entity" title="\"">\"</span>: <label>}, ...]. Points in [y, x] format normalized to 0-1000."</span>
        <span class="token punctuation">}</span>
      <span class="token punctuation">]</span>
    <span class="token punctuation">}</span><span class="token punctuation">]</span>,
    <span class="token string">"generationConfig"</span><span class="token builtin class-name">:</span> <span class="token punctuation">{</span>
      <span class="token string">"temperature"</span><span class="token builtin class-name">:</span> <span class="token number">0.5</span>,
      <span class="token string">"thinkingConfig"</span><span class="token builtin class-name">:</span> <span class="token punctuation">{</span><span class="token string">"thinkingBudget"</span><span class="token builtin class-name">:</span> <span class="token number">0</span><span class="token punctuation">}</span>
    <span class="token punctuation">}</span>
  <span class="token punctuation">}</span>'

prompt <span class="token operator">=</span> <span class="token triple-quoted-string string">"""
Return bounding boxes as a JSON array with labels. Never return masks
or code fencing. Limit to 25 objects. Include as many objects as you
can identify on the table.
If an object is present multiple times, name them according to their
unique characteristic (colors, size, position, etc.).
The format should be: [{"box_2d": [ymin, xmin, ymax, xmax],
"label": <label>}] normalized to 0-1000. Values must be integers.
"""</span>

response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>image_bytes<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span> prompt<span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>
        temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">,</span> thinking_config<span class="token operator">=</span>types<span class="token punctuation">.</span>ThinkingConfig<span class="token punctuation">(</span>thinking_budget<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">)</span>
    <span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

<span class="token keyword">import</span> json

boxes <span class="token operator">=</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"تشخیص </span><span class="token interpolation"><span class="token punctuation">{</span><span class="token builtin">len</span><span class="token punctuation">(</span>boxes<span class="token punctuation">)</span><span class="token punctuation">}</span></span><span class="token string"> شیء"</span></span><span class="token punctuation">)</span>

<span class="token keyword">import</span> cv2

<span class="token comment"># تعریف اشیاء برای ردیابی</span>
queries <span class="token operator">=</span> <span class="token punctuation">[</span>
    <span class="token string">"بلوک قرمز (روی میز)"</span><span class="token punctuation">,</span>
    <span class="token string">"بلوک قرمز (در گیره)"</span><span class="token punctuation">,</span>
    <span class="token string">"سیلندر آبی"</span><span class="token punctuation">,</span>
<span class="token punctuation">]</span>

base_prompt <span class="token operator">=</span> <span class="token string-interpolation"><span class="token string">f"""
Point to the following objects: </span><span class="token interpolation"><span class="token punctuation">{</span><span class="token string">', '</span><span class="token punctuation">.</span>join<span class="token punctuation">(</span>queries<span class="token punctuation">)</span><span class="token punctuation">}</span></span><span class="token string">.
Return JSON: [{{"point": [y, x], "label": <label>}}, ...].
Points in [y, x] format normalized to 0-1000.
If no objects found, return empty list [].
"""</span></span>

<span class="token comment"># بارگذاری ویدیو</span>
video <span class="token operator">=</span> cv2<span class="token punctuation">.</span>VideoCapture<span class="token punctuation">(</span><span class="token string">"robot_manipulation.mp4"</span><span class="token punctuation">)</span>
frame_count <span class="token operator">=</span> <span class="token number">0</span>
tracking_data <span class="token operator">=</span> <span class="token punctuation">[</span><span class="token punctuation">]</span>

<span class="token keyword">while</span> video<span class="token punctuation">.</span>isOpened<span class="token punctuation">(</span><span class="token punctuation">)</span><span class="token punctuation">:</span>
    ret<span class="token punctuation">,</span> frame <span class="token operator">=</span> video<span class="token punctuation">.</span>read<span class="token punctuation">(</span><span class="token punctuation">)</span>
    <span class="token keyword">if</span> <span class="token keyword">not</span> ret<span class="token punctuation">:</span>
        <span class="token keyword">break</span>

    <span class="token comment"># پردازش هر 5 فریم</span>
    <span class="token keyword">if</span> frame_count <span class="token operator">%</span> <span class="token number">5</span> <span class="token operator">==</span> <span class="token number">0</span><span class="token punctuation">:</span>
        <span class="token comment"># رمزگذاری فریم</span>
        _<span class="token punctuation">,</span> <span class="token builtin">buffer</span> <span class="token operator">=</span> cv2<span class="token punctuation">.</span>imencode<span class="token punctuation">(</span><span class="token string">".jpg"</span><span class="token punctuation">,</span> frame<span class="token punctuation">)</span>
        frame_bytes <span class="token operator">=</span> <span class="token builtin">buffer</span><span class="token punctuation">.</span>tobytes<span class="token punctuation">(</span><span class="token punctuation">)</span>

        <span class="token comment"># تشخیص اشیاء</span>
        response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
            model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
            contents<span class="token operator">=</span><span class="token punctuation">[</span>
                types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>frame_bytes<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span>
                base_prompt<span class="token punctuation">,</span>
            <span class="token punctuation">]</span><span class="token punctuation">,</span>
            config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>
                temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">,</span> thinking_config<span class="token operator">=</span>types<span class="token punctuation">.</span>ThinkingConfig<span class="token punctuation">(</span>thinking_budget<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">)</span>
            <span class="token punctuation">)</span><span class="token punctuation">,</span>
        <span class="token punctuation">)</span>

        frame_data <span class="token operator">=</span> <span class="token punctuation">{</span><span class="token string">"frame"</span><span class="token punctuation">:</span> frame_count<span class="token punctuation">,</span> <span class="token string">"objects"</span><span class="token punctuation">:</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>response<span class="token punctuation">.</span>text<span class="token punctuation">)</span><span class="token punctuation">}</span>
        tracking_data<span class="token punctuation">.</span>append<span class="token punctuation">(</span>frame_data<span class="token punctuation">)</span>
        <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"فریم </span><span class="token interpolation"><span class="token punctuation">{</span>frame_count<span class="token punctuation">}</span></span><span class="token string">: </span><span class="token interpolation"><span class="token punctuation">{</span><span class="token builtin">len</span><span class="token punctuation">(</span>frame_data<span class="token punctuation">[</span><span class="token string">'objects'</span><span class="token punctuation">]</span><span class="token punctuation">)</span><span class="token punctuation">}</span></span><span class="token string"> شیء تشخیص داده شد"</span></span><span class="token punctuation">)</span>

    frame_count <span class="token operator">+=</span> <span class="token number">1</span>

video<span class="token punctuation">.</span>release<span class="token punctuation">(</span><span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"پردازش </span><span class="token interpolation"><span class="token punctuation">{</span><span class="token builtin">len</span><span class="token punctuation">(</span>tracking_data<span class="token punctuation">)</span><span class="token punctuation">}</span></span><span class="token string"> فریم"</span></span><span class="token punctuation">)</span>

prompt <span class="token operator">=</span> <span class="token triple-quoted-string string">"""
Place a point on the red block, then 15 points for the trajectory of
moving the red block to the top of the container on the left.
The points should be labeled by order of the trajectory, from '0'
(start point) to '15' (final point).
Return JSON: [{"point": [y, x], "label": <label>}, ...].
Points in [y, x] format normalized to 0-1000.
"""</span>

response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>image_bytes<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span> prompt<span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

trajectory <span class="token operator">=</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"مسیر تولید شده با </span><span class="token interpolation"><span class="token punctuation">{</span><span class="token builtin">len</span><span class="token punctuation">(</span>trajectory<span class="token punctuation">)</span><span class="token punctuation">}</span></span><span class="token string"> نقطه میانی"</span></span><span class="token punctuation">)</span>


<span class="token comment"># تبدیل مختصات نرمال شده به مختصات فضای کاری ربات</span>
<span class="token keyword">def</span> <span class="token function">normalize_to_robot_coords</span><span class="token punctuation">(</span>point<span class="token punctuation">,</span> workspace_dims<span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token triple-quoted-string string">"""تبدیل مختصات نرمال (0-1000) به مختصات فضای کاری ربات"""</span>
    y_norm<span class="token punctuation">,</span> x_norm <span class="token operator">=</span> point
    x_robot <span class="token operator">=</span> <span class="token punctuation">(</span>x_norm <span class="token operator">/</span> <span class="token number">1000.0</span><span class="token punctuation">)</span> <span class="token operator">*</span> workspace_dims<span class="token punctuation">[</span><span class="token string">"width"</span><span class="token punctuation">]</span> <span class="token operator">+</span> workspace_dims<span class="token punctuation">[</span><span class="token string">"x_min"</span><span class="token punctuation">]</span>
    y_robot <span class="token operator">=</span> <span class="token punctuation">(</span>y_norm <span class="token operator">/</span> <span class="token number">1000.0</span><span class="token punctuation">)</span> <span class="token operator">*</span> workspace_dims<span class="token punctuation">[</span><span class="token string">"height"</span><span class="token punctuation">]</span> <span class="token operator">+</span> workspace_dims<span class="token punctuation">[</span><span class="token string">"y_min"</span><span class="token punctuation">]</span>
    <span class="token keyword">return</span> <span class="token punctuation">(</span>x_robot<span class="token punctuation">,</span> y_robot<span class="token punctuation">)</span>


workspace <span class="token operator">=</span> <span class="token punctuation">{</span><span class="token string">"x_min"</span><span class="token punctuation">:</span> <span class="token operator">-</span><span class="token number">0.5</span><span class="token punctuation">,</span> <span class="token string">"width"</span><span class="token punctuation">:</span> <span class="token number">1.0</span><span class="token punctuation">,</span> <span class="token string">"y_min"</span><span class="token punctuation">:</span> <span class="token operator">-</span><span class="token number">0.5</span><span class="token punctuation">,</span> <span class="token string">"height"</span><span class="token punctuation">:</span> <span class="token number">1.0</span><span class="token punctuation">}</span>

robot_trajectory <span class="token operator">=</span> <span class="token punctuation">[</span><span class="token punctuation">]</span>
<span class="token keyword">for</span> waypoint <span class="token keyword">in</span> trajectory<span class="token punctuation">:</span>
    robot_coords <span class="token operator">=</span> normalize_to_robot_coords<span class="token punctuation">(</span>waypoint<span class="token punctuation">[</span><span class="token string">"point"</span><span class="token punctuation">]</span><span class="token punctuation">,</span> workspace<span class="token punctuation">)</span>
    robot_trajectory<span class="token punctuation">.</span>append<span class="token punctuation">(</span><span class="token punctuation">{</span><span class="token string">"label"</span><span class="token punctuation">:</span> waypoint<span class="token punctuation">[</span><span class="token string">"label"</span><span class="token punctuation">]</span><span class="token punctuation">,</span> <span class="token string">"coords"</span><span class="token punctuation">:</span> robot_coords<span class="token punctuation">}</span><span class="token punctuation">)</span>
    <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"نقطه میانی </span><span class="token interpolation"><span class="token punctuation">{</span>waypoint<span class="token punctuation">[</span><span class="token string">'label'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">: </span><span class="token interpolation"><span class="token punctuation">{</span>robot_coords<span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>

prompt <span class="token operator">=</span> <span class="token triple-quoted-string string">"""
Point to the object that I need to remove to make room for my laptop.
Return JSON: [{"point": [y, x], "label": <label>}, ...].
Points in [y, x] format normalized to 0-1000.
"""</span>

response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>desk_image<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span> prompt<span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>
        temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">,</span> thinking_config<span class="token operator">=</span>types<span class="token punctuation">.</span>ThinkingConfig<span class="token punctuation">(</span>thinking_budget<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">)</span>
    <span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

result <span class="token operator">=</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"شیء برای حذف: </span><span class="token interpolation"><span class="token punctuation">{</span>result<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token string">'label'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"موقعیت: </span><span class="token interpolation"><span class="token punctuation">{</span>result<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token string">'point'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>

prompt <span class="token operator">=</span> <span class="token triple-quoted-string string">"""
Explain how to pack the lunch box and lunch bag. Point to each
object you refer to. Each point format:
[{"point": [y, x], "label": <object_name>}]
Coordinates normalized to 0-1000.
"""</span>

response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>lunch_image<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span> prompt<span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>
        temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">,</span> thinking_config<span class="token operator">=</span>types<span class="token punctuation">.</span>ThinkingConfig<span class="token punctuation">(</span>thinking_budget<span class="token operator">=</span><span class="token number">0</span><span class="token punctuation">)</span>
    <span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

<span class="token keyword">print</span><span class="token punctuation">(</span>response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>

<span class="token comment"># تعریف API ربات شبیه‌سازی شده</span>
<span class="token keyword">def</span> <span class="token function">move</span><span class="token punctuation">(</span>x<span class="token punctuation">,</span> y<span class="token punctuation">,</span> high<span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token triple-quoted-string string">"""حرکت بازو به مختصات. high=True بازو را بالای صحنه بلند می‌کند."""</span>
    <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"حرکت به: x=</span><span class="token interpolation"><span class="token punctuation">{</span>x<span class="token punctuation">}</span></span><span class="token string">, y=</span><span class="token interpolation"><span class="token punctuation">{</span>y<span class="token punctuation">}</span></span><span class="token string">, z=</span><span class="token interpolation"><span class="token punctuation">{</span><span class="token string">'بالا'</span> <span class="token keyword">if</span> high <span class="token keyword">else</span> <span class="token string">'پایین'</span><span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>


<span class="token keyword">def</span> <span class="token function">setGripperState</span><span class="token punctuation">(</span>opened<span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token triple-quoted-string string">"""باز یا بسته کردن گیره."""</span>
    <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string">"باز کردن گیره"</span> <span class="token keyword">if</span> opened <span class="token keyword">else</span> <span class="token string">"بستن گیره"</span><span class="token punctuation">)</span>


<span class="token keyword">def</span> <span class="token function">returnToOrigin</span><span class="token punctuation">(</span><span class="token punctuation">)</span><span class="token punctuation">:</span>
    <span class="token triple-quoted-string string">"""بازگشت به حالت اولیه."""</span>
    <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string">"بازگشت به مبدا"</span><span class="token punctuation">)</span>


<span class="token comment"># ابتدا، مکان اشیاء را پیدا کنید</span>
locate_prompt <span class="token operator">=</span> <span class="token triple-quoted-string string">"""
Locate and point to the blue block and the orange bowl.
Return JSON: [{"point": [y, x], "label": <label>}, ...].
Points in [y, x] format normalized to 0-1000.
"""</span>

locate_response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>
        types<span class="token punctuation">.</span>Part<span class="token punctuation">.</span>from_bytes<span class="token punctuation">(</span>data<span class="token operator">=</span>scene_image<span class="token punctuation">,</span> mime_type<span class="token operator">=</span><span class="token string">"image/jpeg"</span><span class="token punctuation">)</span><span class="token punctuation">,</span>
        locate_prompt<span class="token punctuation">,</span>
    <span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

objects <span class="token operator">=</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>locate_response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>
block <span class="token operator">=</span> <span class="token builtin">next</span><span class="token punctuation">(</span>obj <span class="token keyword">for</span> obj <span class="token keyword">in</span> objects <span class="token keyword">if</span> <span class="token string">"block"</span> <span class="token keyword">in</span> obj<span class="token punctuation">[</span><span class="token string">"label"</span><span class="token punctuation">]</span><span class="token punctuation">.</span>lower<span class="token punctuation">(</span><span class="token punctuation">)</span><span class="token punctuation">)</span>
bowl <span class="token operator">=</span> <span class="token builtin">next</span><span class="token punctuation">(</span>obj <span class="token keyword">for</span> obj <span class="token keyword">in</span> objects <span class="token keyword">if</span> <span class="token string">"bowl"</span> <span class="token keyword">in</span> obj<span class="token punctuation">[</span><span class="token string">"label"</span><span class="token punctuation">]</span><span class="token punctuation">.</span>lower<span class="token punctuation">(</span><span class="token punctuation">)</span><span class="token punctuation">)</span>

<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"بلوک در: </span><span class="token interpolation"><span class="token punctuation">{</span>block<span class="token punctuation">[</span><span class="token string">'point'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string-interpolation"><span class="token string">f"کاسه در: </span><span class="token interpolation"><span class="token punctuation">{</span>bowl<span class="token punctuation">[</span><span class="token string">'point'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">"</span></span><span class="token punctuation">)</span>

<span class="token comment"># حالا عملیات برداشتن و قرار دادن را هماهنگ کنید</span>
robot_origin <span class="token operator">=</span> <span class="token punctuation">[</span><span class="token number">500</span><span class="token punctuation">,</span> <span class="token number">500</span><span class="token punctuation">]</span>  <span class="token comment"># مختصات نرمال شده</span>
block_relative <span class="token operator">=</span> <span class="token punctuation">[</span>
    block<span class="token punctuation">[</span><span class="token string">"point"</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span> <span class="token operator">-</span> robot_origin<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span><span class="token punctuation">,</span>
    block<span class="token punctuation">[</span><span class="token string">"point"</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token number">1</span><span class="token punctuation">]</span> <span class="token operator">-</span> robot_origin<span class="token punctuation">[</span><span class="token number">1</span><span class="token punctuation">]</span><span class="token punctuation">,</span>
<span class="token punctuation">]</span>
bowl_relative <span class="token operator">=</span> <span class="token punctuation">[</span>bowl<span class="token punctuation">[</span><span class="token string">"point"</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span> <span class="token operator">-</span> robot_origin<span class="token punctuation">[</span><span class="token number">0</span><span class="token punctuation">]</span><span class="token punctuation">,</span> bowl<span class="token punctuation">[</span><span class="token string">"point"</span><span class="token punctuation">]</span><span class="token punctuation">[</span><span class="token number">1</span><span class="token punctuation">]</span> <span class="token operator">-</span> robot_origin<span class="token punctuation">[</span><span class="token number">1</span><span class="token punctuation">]</span><span class="token punctuation">]</span>

orchestrate_prompt <span class="token operator">=</span> <span class="token string-interpolation"><span class="token string">f"""
You are a robotic arm with six degrees-of-freedom. You have these functions:

def move(x, y, high):
  # Moves arm to coordinates. high=True lifts arm above scene.

def setGripperState(opened):
  # Opens gripper if opened=True, closes if False

def returnToOrigin():
  # Returns robot to initial state

Origin point is at normalized </span><span class="token interpolation"><span class="token punctuation">{</span>robot_origin<span class="token punctuation">}</span></span><span class="token string">.
Perform pick and place: pick up blue block at </span><span class="token interpolation"><span class="token punctuation">{</span>block<span class="token punctuation">[</span><span class="token string">'point'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">
(relative: </span><span class="token interpolation"><span class="token punctuation">{</span>block_relative<span class="token punctuation">}</span></span><span class="token string">) and place in orange bowl at </span><span class="token interpolation"><span class="token punctuation">{</span>bowl<span class="token punctuation">[</span><span class="token string">'point'</span><span class="token punctuation">]</span><span class="token punctuation">}</span></span><span class="token string">
(relative: </span><span class="token interpolation"><span class="token punctuation">{</span>bowl_relative<span class="token punctuation">}</span></span><span class="token string">).

Provide sequence of function calls as JSON list:
[{{"function": <name>, "args": [<args>]}}, ...]
Include your reasoning before the JSON.
"""</span></span>

orchestrate_response <span class="token operator">=</span> client<span class="token punctuation">.</span>models<span class="token punctuation">.</span>generate_content<span class="token punctuation">(</span>
    model<span class="token operator">=</span>MODEL_ID<span class="token punctuation">,</span>
    contents<span class="token operator">=</span><span class="token punctuation">[</span>orchestrate_prompt<span class="token punctuation">]</span><span class="token punctuation">,</span>
    config<span class="token operator">=</span>types<span class="token punctuation">.</span>GenerateContentConfig<span class="token punctuation">(</span>temperature<span class="token operator">=</span><span class="token number">0.5</span><span class="token punctuation">)</span><span class="token punctuation">,</span>
<span class="token punctuation">)</span>

<span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string">"برنامه مدل:"</span><span class="token punctuation">)</span>
<span class="token keyword">print</span><span class="token punctuation">(</span>orchestrate_response<span class="token punctuation">.</span>text<span class="token punctuation">)</span>

<span class="token comment"># تجزیه و اجرای فراخوانی‌های تابع</span>
<span class="token keyword">import</span> re

json_match <span class="token operator">=</span> re<span class="token punctuation">.</span>search<span class="token punctuation">(</span><span class="token string">r"\[.*\]"</span><span class="token punctuation">,</span> orchestrate_response<span class="token punctuation">.</span>text<span class="token punctuation">,</span> re<span class="token punctuation">.</span>DOTALL<span class="token punctuation">)</span>
<span class="token keyword">if</span> json_match<span class="token punctuation">:</span>
    function_calls <span class="token operator">=</span> json<span class="token punctuation">.</span>loads<span class="token punctuation">(</span>json_match<span class="token punctuation">.</span>group<span class="token punctuation">(</span><span class="token punctuation">)</span><span class="token punctuation">)</span>
    <span class="token keyword">print</span><span class="token punctuation">(</span><span class="token string">"\nدر حال اجرا:"</span><span class="token punctuation">)</span>
    <span class="token keyword">for</span> call <span class="token keyword">in</span> function_calls<span class="token punctuation">:</span>
        func_name <span class="token operator">=</span> call<span class="token punctuation">[</span><span class="token string">"function"</span><span class="token punctuation">]</span>
        args <span class="token operator">=</span> call<span class="token punctuation">[</span><span class="token string">"args"</span><span class="token punctuation">]</span>
        <span class="token keyword">if</span> func_name <span class="token operator">==</span> <span class="token string">"move"</span><span class="token punctuation">:</span>
            move<span class="token punctuation">(</span><span class="token operator">*</span>args<span class="token punctuation">)</span>
        <span class="token keyword">elif</span> func_name <span class="token operator">==</span> <span class="token string">"setGripperState"</span><span class="token punctuation">:</span>
            setGripperState<span class="token punctuation">(</span><span class="token operator">*</span>args<span class="token punctuation">)</span>
        <span class="token keyword">elif</span> func_name <span class="token operator">==</span> <span class="token string">"returnToOrigin"</span><span class="token punctuation">:</span>
            returnToOrigin<span class="token punctuation">(</span><span class="token punctuation">)</span>