Google DeepMind 发布面向机器人的 Gemini 4 Argon

2026年10月13日

Gemini 4 Argon 在物理理解、代码生成和视觉语义推理方面实现了跨越式提升，为整个机器人技术栈解锁了新的工作流程，并加速了迈向具身 AGI 的进程。

你可以在 具身推理（ERQA）基准上看到这一能力飞跃。该基准测试模型在 3D 环境中理解并解决物理和空间问题的能力。仅凭推理，Argon 就取得了 87% 的成绩，超过了其他所有前沿模型。

作为通用物理推理模型，Argon 可应用于整个机器人生命周期。在本文中，我们将分享 Argon 在四个关键领域的代表性提示词：

• 机器人控制 → 使用 MuJoCo 进行移动操作练习、零样本机器人控制，以及对视频世界模型进行闭环操控。
• Real2Sim → 根据机器人摄像头画面、以自我为中心的人类视频或手绘纸面草图，重建经过物理校准的仿真孪生体。
• 物理理解 → 具备智能体能力的多模态事后标注，包括对灵巧性的详细理解、密集价值函数和进度估计。
• 研究与自我改进 → 通过仿真改进行动，并在硬件上进行迭代式自我改进。

使用 Argon 解锁机器人控制

1. Spot 移动操作：使用 MuJoCo 练习将垃圾投入垃圾桶

Argon 控制 Boston Dynamics Spot® 四足机器人（12-DoF 腿部 + 6-DoF 机械臂）学习一项新技能：踩下脚踏式垃圾桶的踏板，将汽水罐扔进去。Argon 创建脚踏式垃圾桶的铰接式 MuJoCo 数字孪生体，在踩下实体踏板并将汽水罐丢入打开的垃圾桶之前，先练习相关动作。

抱歉，你的浏览器不支持播放此视频

Spot 移动操作——从真实环境到仿真再回到真实环境的脚踏式垃圾桶投放：Spot 拿起汽水罐，检查腕部摄像头，在 MuJoCo 中练习三腿三脚架平衡，踩下踏板，确认手部位于打开的垃圾桶中央，然后将汽水罐丢入其中。

提示词：

There's a Diet Coke can behind you and another on the black table on your left.
Use the spot-mobile-manipulation skill to pick them up and trash them in the black foot-pedal bin.
Before stepping down on the pedal, verify in the front cameras that your front-right foot target is centered on the pedal, and use a 2-agent verification swarm (foot-verifier and hand-verifier) to verify both the foot-pedal step and the 3D hand-in-bin alignment before releasing.

纯文本

已复制

2. 零样本控制

Argon 控制一个简单的机械臂，将倒下的马克杯扶正，并把揉皱的餐巾放进去。Argon 借助视觉坐标辅助系统和动作原语，通过多轮交互完成这项任务。

抱歉，你的浏览器不支持播放此视频

左：第三人称真实机器人视角；右：经过校准的腕部摄像头，画面中投影了间隔 2 cm 的世界坐标和闭合夹爪的瞄准标记。

提示词：

[Attached: wrist_cam.png, obs.json]

Task: Get the robot arm to reorient the mug upright and put the napkin inside the mug.

## Execution Discipline
- Simply output 1 sentence observation followed immediately by your `curl` command (with `# Plan:` on Turn 1 and `# Thought:` on every turn). Everything needed is in `obs.json` and the calibrated reticle/grid drawn on `wrist_cam.png`.

## Visual Coordinate Reference (Annotated on wrist_cam.png)
- `+X` (UP in image): Away from robot base | `-X` (DOWN): Towards robot base
- `+Y` (LEFT in image): World left         | `-Y` (RIGHT): World right
- `+Z`: Height above table                | `Yellow numbers`: World (X, Y) in centimetres (2 cm grid at `grid_z`)
- `Cyan + crosshair`: Current 3D pinch center
- `Cyan [ + ] brackets`: Current jaw opening and closing line
- `Green line`: Current object long-axis orientation (`yaw_deg`: 0° = vertical along +X, 90° = horizontal along +Y)
- `pitch_deg = 90°`: Straight down (-Z) top-down grasp/place; `pitch_deg < 60°`: Tilted coaxial approach along finger axis

## Closed-Loop Turn Workflow
1. Inspect current state & view `wrist_cam.png`:
   `curl -s http://127.0.0.1:8081/obs` (returns `ee_pos`, `holding`, `status`, `workspace`, `wrist_cam_path`).
2. Select VLM call granularity upfront on Turn 1:
   - Coarse (2–4 calls): `look_at` -> `grasp` -> `place` -> `done`
   - Pre-Grasp Hover Verify (4–7 calls): `look_at` -> `goto(z=0.04)` (verify cyan `[ + ]` brackets straddle target) -> `grasp` -> `place_until_contact` -> `done`
   - Fine Visual Servoing (8–20 calls): `look_at` -> `grasp` -> pre-pose `goto` -> closed-loop `move_by` / `move_until_contact` -> `done`
3. Send motion primitives (`POST /plan`) and verify visually after every turn in the updated `wrist_cam.png`. If `status == "grasp_failed"`, call `robot.look_at(...)` to re-observe and recover closed-loop.

## Code-as-Policies Primitives (`POST http://127.0.0.1:8081/plan`)
robot.look_at(x=0.35, y=0.13, z=0.14, grid_z=0.015)
x, y = robot.deproject(u=0.52, v=0.48, z_plane=0.015)
robot.grasp(x=x, y=y, z=0.015, yaw_deg=-35, pitch_deg=90)
robot.place(x=0.24, y=0.06, z=0.075)
robot.place_until_contact(x=0.24, y=0.06, z=0.015, max_lag=0.012, relief=0.002)
robot.goto(x=0.28, y=0.05, z=0.04, arc_z=0.14, pitch_deg=25, azimuth_deg=90, roll_deg=0, speed=0.10)
robot.move_by(forward=0.025, dz=-0.002, dpitch_deg=-3, droll_deg=5, speed=0.04)
robot.move_until_contact(dz=-0.06, speed=0.03, max_lag=0.012, relief=0.002)
robot.open_gripper() / robot.close_gripper() / robot.home() / robot.done()

纯文本

已复制

3. 全向控制

Argon 在 Gemini Omni World Model 中操控机器人，将五个散落的 Scrabble 字母方块（G, R, A, N, O）重新排列成一行，拼出 A - R - G - O - N。Argon 并不指定一个静态提示词，而是在闭环视觉循环中进行操控。它编写带时间戳的动作提示词，检查中间 rollout 的关键帧，并不断优化提示词。

抱歉，你的浏览器不支持播放此视频

迭代式视频世界模型灵巧手操控：33 步操控视频 rollout，拼出 A-R-G-O-N。

提示词：

Goal: {task_goal}

You are steering a video world model (Gemini Omni) in an iterative visual feedback loop. Inspect the attached frames from the latest rollout and write the next timestamped action prompt for the video world model:
- Break complex motions into one simple, single-direction action at a time.
- Keep inactive hands retracted and avoid blocking the camera's view of objects in the workspace.
- Explicitly state which areas of the workspace become empty after an object moves.
- Use [MM:SS.mmm] timestamps for each action beat and wrap your output in <omni_prompt> tags.

Example format (sample timestamps only, not the full prompt):
<omni_prompt>
[00:00.000] Character A's left hand rests at the side while the right hand reaches toward the target object.
[00:02.500] Character A's right fingertips push the object smoothly into the center of the table, leaving its starting area empty.
[00:05.000] Character A's right hand opens and retracts to the side so all placed objects remain in clear view of the camera.
</omni_prompt>

纯文本

已复制

使用 Argon 解锁 Real2Sim

1. 从机器人视角到仿真

Argon 查看真实机器人的遥操作片段并构建数字孪生；它获取或创建外观和行为都与真实物体相符的仿真资产，使其在与灵巧手接触时表现一致。Argon 通过迭代式视觉推理循环调整资产和场景，使仿真视觉效果及接触事件与真实录制数据中的相应内容相匹配。

抱歉，您的浏览器不支持播放此视频

左侧（真实视角）：同步的真实机器人摄像头画面；右侧（仿真视角）：由 Argon 重建并校准的 3D MuJoCo 场景。

提示：

[Attached: robot_rollout_episode]

Build a MuJoCo scene of the robot and reconstructed task objects. When each episode is replayed frame by frame, by driving the robot joints toward the logged observation.state with joint-space PD.

Iterate using two verification objectives:
(1) Visual match: MuJoCo renders from the head_left, left_wrist, and right_wrist cameras line up with, and look like, those real observations.
(2) Physical match: the simulated contact force/torque on each Sharpa fingertip, in the sensor frame, matches the logged tactile_force in timing, magnitude and direction.

What you can tune happens before simulation starts:
(1) Static object assets: realistic geometry, collision shapes, mass, inertia, friction and joints, with appearance close to the real objects, plus the initial object poses at t = 0.
(2) Offline-planned robot commands, changed only in frames where the real fingertips are near an object. Each command must stay within 0.55 rad (arm joints) and 0.65 rad (hand joints) of the logged teleop trajectory. Everywhere else, replay the logged trajectory unchanged.

No sim cheating: task objects may move only through real MuJoCo contact with the robot hands and the environment (mj_step).
Not allowed after t = 0: changing MjModel parameters; applying external forces or torques to objects; welds or equality constraints; feedforward torques beyond joint-space PD; setting or rolling back object qpos/qvel; disabling or excluding collisions; reporting contact forces that don't come from the MuJoCo solver (for example, copying the logged tactile readings).

Immediately present the reconstructed scene and rollout results as an interactive HTML session viewer with one page per episode, showing:
- At the bottom: A synced timeline (play, scrub, speed control) that drives every panel.
- On the left panel: An interactive 3D Blender scene showing the simulated robot and objects, showing contact points and force arrows as replay progresses.
- On the right panel: Real vs. sim camera views side by side for all 3 cameras. Real vs. sim finger force visualizations.

纯文本

已复制

2. 从人类视角到仿真

给定一段关于操作任务的第一人称人类视频，Argon 会识别与任务相关的物体，将粗略的边界框替换为适用于凹形容器的复合凸分解网格，并在 MuJoCo 接触物理环境中，将人手轨迹重定向到一台拥有 76-DoF 双臂类人机器人的轨迹上，该机器人配备 22-DoF 五指灵巧手。

抱歉，您的浏览器不支持播放此视频

从人类视角到仿真——从第一人称人类视频到灵巧类人机器人孪生体：左侧显示输入的第一人称人类 RGB 视频；右侧显示同步的 500 Hz MuJoCo + Blender Cycles 多视角仿真运行过程。

提示：

[Attached: egocentric_human_video.mp4, candidate_sim_views.png, hand_3d_keypoints.json]

You are building a physically solvable 500 Hz MuJoCo manipulation environment (`scene.xml`) and studio Blender render that faithfully reproduces the tabletop task shown in `egocentric_human_video.mp4` on a bimanual humanoid robot with 22-DoF dexterous five-fingered hands:

1. Scene Reconstruction & Camera Calibration:
   - Identify only the manipulation-relevant objects (the grasped target object and the receiving container or landmark) and triangulate their 3D tabletop coordinates from the egocentric camera intrinsics/extrinsics.
   - Replace crude bounding boxes with compound convex-decomposed 3D meshes so concave containers (nested bowls, recessed wooden trays, open bins) support physical interior placement.

2. 500 Hz Dexterous Motion-Fitting Stress Test:
   - Retarget the 3D human wrist and finger keypoints to the humanoid's 7-DoF arm and 22-DoF hand joints, and step the trajectory in MuJoCo at 500 Hz (dt = 0.002s) with zero kinematic weld constraints.
   - Verify opposing thumb-and-finger force closure, clean object lift (> 2 cm), and stable final placement inside or beside the target landmark.

纯文本

已复制

3. 从草图到仿真

借助 Argon，开发者可以拍摄一张粗略的纸笔草图——图中展示 Apollo 2 类人机器人沿抛物线轨迹投掷球，使桌面上由箱子堆成的 3-2-1 结构倒塌——Argon 随后会自动构建 GPU 加速的 mujoco_warp 场景，合成平衡的 76-DoF 全身及五指投掷轨迹，在闭环物理环境中验证 6/6 的目标击倒效果，并导出 Blender 4.5 Cycles 渲染图。

抱歉，您的浏览器不支持播放此视频

从草图到仿真——手绘草图到灵巧类人机器人轨迹与 Cycles 渲染图。

提示：

[Attached: hand_drawn_throw_sketch.jpg]

Translate this hand-drawn sketch into a physically verified humanoid simulation in `mujoco_warp` and a raytraced Blender Cycles video:

1. Scene & Kinematic Setup from Sketch:
   - Parse the spatial layout from `hand_drawn_throw_sketch.jpg`: place the 76-DoF humanoid (32-DoF body equipped with two 22-DoF five-finger dexterous hands) at the origin facing a table holding a 3-2-1 pyramid of 6 rigid target boxes, with a projectile sphere grasped in the right five-finger hand.

2. Whole-Body Balance & Dexterous Throw Synthesis:
   - Synthesize a 50 Hz whole-body throwing trajectory (140 steps, 2.80 s) across four phases: (1) stable stance & 5-finger power grasp closure, (2) torso/shoulder wind-up with ankle-hip Center-of-Mass compensation, (3) forward whipping acceleration, and (4) synchronized 22-DoF five-finger opening at peak hand velocity so the ball launches on a ballistic arc into the pyramid's second-tier center of mass.
   - Verify in `mujoco_warp` contact physics that the humanoid maintains upright balance (< 3 cm pelvis drift) and knocks down all 6 target boxes.

纯文本

已复制

借助 Argon 解锁物理理解能力

1. 事后标注与数据质量

Argon 对灵巧操作具备精确的物理理解能力，能够从视频中提供详细的分割结果和文本标注。Argon 可以识别暂停、不稳定抓取和失误等低质量操作片段。给定一段俯拍视频，其中一台双臂 ALOHA 2 机器人用 136 秒将一张方形纸折成一只折纸狐狸，Argon 会生成时间对齐的四级事后标注层级：任务目标（G1）、6 个连续的语义子任务（G2）、17 个接触级原子操作（G3），以及自动化质量控制（QC）重新加权机制，在失误和重试期间保持任务进度不变。

抱歉，您的浏览器不支持播放此视频

Argon 多粒度事后分析与遥测 HUD。

提示：

[Attached: Aloha2_Origami_20250309.mp4]

Produce a time-aligned multi-granularity hindsight annotation JSON with four levels:
1. G1_episode_instruction: Concise imperative sentence describing the overall manipulation goal.
2. G2_subtasks: Contiguous semantic subtask phases specifying which arm(s) anchor vs. fold/rotate the object.
3. G3_atomic_primitives: Fine-grained contact-level primitives (3–15s) aligned with grasp closure, crease sweeps, release, or retry attempts.
4. quality_control_and_reweighting: Hold task_progress_pct flat during non-progressing slip/delamination retry loops and downweight them (0.15x–0.25x) for Behavior Cloning.

纯文本

已复制

2. 找出异类球

为了用机器人重现经典的 12 球天平谜题，Argon 构建了一个双臂 Franka FR3 MuJoCo 环境，其中有 12 个外观完全相同的台球（其中一个暗中更重或更轻）以及一个带关节的 Roberval 双盘天平。随后，它编写了一个仅依靠视觉的闭环求解器：通过立体 RGB 摄像头对球进行三角定位，读取每次称量后的天平倾斜情况，并在 3 次称量内将异类球确定到较重（红色）或较轻（蓝色）托盘中。

抱歉，您的浏览器不支持播放此视频

找出异色球（仅视觉闭环求解器）：双臂 Franka FR3 在金色/青绿色托盘上称量候选球，读取立体视觉秤的倾斜情况，并通过 3 次称量将确认较重的异色球单独放入红色托盘。

提示：

# Prompt 1 — Physics-Accurate MuJoCo Environment & Benchmark Generation
Make a MuJoCo environment where a dual-arm robot (two 7-DoF Franka FR3 arms with parallel-jaw grippers on a shared tabletop) is tasked to solve the classic 12-ball balance scale weight puzzle from https://www.mathsisfun.com/pool_balls_solution.html.
1. Create 12 visually indistinguishable pool balls in a 2x6 starting rack (11 standard 150 g, 1 odd ball either 270 g or 30 g).
2. Build a functional two-pan Roberval balance scale at the center of the table with parallelogram equality constraints (`mjEQ_JOINT`) so the Left (Gold) and Right (Teal) pans stay horizontal, plus Red (Heavy) and Blue (Light) submission trays.
3. Track distinct settled weighings (`weighing_count`) and reward `1.00` for isolating the true odd ball in the correct Heavy/Light tray within <= 3 weighings.

# Prompt 2 — Closed-Loop Vision-Only Solver
Implement an independent closed-loop policy relying strictly on stereo RGB camera inputs (`head_camera_left_rgb_image`, `head_camera_right_rgb_image`) and robot proprioception (`add_privileged_task_observations=False`):
1. Segment the 12 balls, Gold/Teal pans, and Red/Blue trays in both stereo views and triangulate their 3D coordinates.
2. Visually read the balance scale tilt after each weighing from stereo vertical pixel disparity (+1 = left heavier, -1 = right heavier, 0 = balanced).
3. Rotate the gripper wrist yaw by 90 degrees when grasping/placing inside the narrow scale pans to clear the raised rims, and isolate the odd ball into the Red or Blue tray within <= 3 weighings.

纯文本

已复制

3. 价值函数

Argon 是一种适用于机器人任务的通用价值函数。在一段来自人类 UMI 数据集、时长 60 秒的双手环形物体分拣任务中，Argon 会在接触状态转换前后，反复检查头部、左夹爪和右夹爪摄像头的同步画面，生成密集的价值函数/任务进度预测，并将轨迹分解为 28 个片段，每个片段具有正向、中性或负向的进度斜率。

抱歉，您的浏览器不支持播放此视频

价值函数：与动画任务进度曲线和双手操作片段标注同步的 UMI 三摄像头（头部、左夹爪、右夹爪）回放。

提示：

[Attached: umi_head.mp4, umi_left_gripper.mp4, umi_right_gripper.mp4]

Look at 3 videos of human wearing a gripper performing activities from head, left, right gripper views. Look at the videos thoroughly and give a detailed plot of progress at the task (with positive, neutral, negative slope), identifying important segments in the video and labeling the activities. Look at individual frames or subsegments of the video if necessary to get a better understanding.

纯文本

已复制

借助 Argon 解锁研究与自我改进

1. 通过仿真改进动作

Argon 可在仿真中生成新数据，以提升通用视觉-语言-动作（VLA）模型的性能。仅从一张真实桌面的 RGB 照片出发，Argon 就能自主构建经过校准的 MuJoCo 仿真孪生体、定义新任务，并合成闭环运动生成器，在仿真中采集 4,500+ 个经过领域随机化的回合——使真实世界实体机器人的成功率从 56% 提升至 75%，完成时间从 35s 缩短至 24s。

抱歉，您的浏览器不支持播放此视频

通过仿真改进动作：提升真实世界评测基准（成功率从 56% → 75%，完成时间从 35s → 24s）。

提示：

[Attached: real_tabletop_photo.jpg]

You are given a single overhead RGB camera photo (`real_tabletop_photo.jpg`) of a bimanual robot workspace (two 7-DoF Franka FR3 arms with parallel-jaw grippers and 4 RGB cameras) facing a table arranged with photobooth party props.
1. Reconstruct procedural 3D meshes and compound MuJoCo collision primitives for all 9 party props matching their real-world dimensions, colors, and resting poses, including spring-hinged deformable crown panels for the soft plush cap.
2. Define a 77-task single-arm and bimanual curriculum (pick-and-place, uprighting, directional pushing, mid-air handovers, bimanual scoop-and-push, and pipelined table clearing).
3. Synthesize a closed-loop IK trajectory generator across ±10 cm pose randomization, verify `reward == 1.0` in MuJoCo physics, and export 4,500+ episodes for VLA fine-tuning.

纯文本

已复制

2. 视觉抓取与放置的自我改进

在一台简单的 6 自由度机械臂上，Argon 通过试验和自我反思，迭代提升抓取与放置任务的执行速度和放置精度。

抱歉，您的浏览器不支持播放此视频

SO-101 视觉引导抓取与放置——跨第 1 → 2 → 3 次迭代的硬件自我改进：第 1 次迭代（58.9s）、第 2 次迭代（22.9s，放下物体时擦到盒沿）和第 3 次迭代（17.4s，精准放在正中央）。

提示：

# Stage 1 — Measure Objects & Dry-Run Plan (No Hardware Motion)
Stage 1: measure objects — find the objects and plan the whole motion. The arm does not move in this stage.

A. I put a LEGO brick and a box on the table in front of the arm, and the wooden box to one side. Find the brick and the wooden box in the image. And also in the real world. Show this in an annotated image.

B. A dry-run plan (no hardware). Write the planner to pick and place the lego block. Print the plan table, draw all poses onto the overhead image, and stop. Do not move the robot.
Report with evidence.

# Stage 2 — Pre-Pick Hover & Wrist-Camera Verification
Stage 2: Put the gripper above the brick without touching it, and verify with the wrist camera that the arm and the vision agree. Please record executions with both cameras (compressed video is OK).

# Stage 3 (Autonomous Pick-and-Place → Self-Improvement → Optimization):
1. Execute autonomously: Turn the Stage 1 dry run into a real grasp, lift, and place into the wooden box. Run it yourself subject to safety rules from AGENTS.md.

2. Self-improve: When an attempt fails, diagnose it yourself. Review your own logs and wrist/overhead footage, find the root cause, check the fix in a dry run, and retry.

3. Optimize on feedback: Once it succeeds, make it at least 2× faster within safety limits (rule R5) unless I authorize more, or give feedback.

纯文本

已复制

3. 在 MuJoCo WebAssembly 中通过单条提示词复现机器人论文

对于每篇论文，Argon 只需一条提示词，就能阅读机器人控制论文及其公开的检查点，将物理模型和闭环控制器（从 DeepLoco 中的 7 网络分层运动控制，到 HUSKY 中的 50 Hz PPO 策略网络，再到 Whole-Body MPC 中的 Web Worker iLQR）移植到独立的 MuJoCo WebAssembly 浏览器实验室中，并在无头 Chromium 中进行验证。

抱歉，您的浏览器不支持播放此视频

五篇经典腿式机器人与人形机器人论文在浏览器中实时运行（MuJoCo WASM）：机器人足球 OP3 1v1、DeepLoco、大规模并行深度强化学习 ANYmal C、Whole-Body MPC Go2，以及 HUSKY G1 滑板（来自 Argon 复现的 68 个交互式浏览器沙盒）。

提示：

[Attached: paper.pdf, optional author_checkpoints/ or urdf_assets/]

Reproduce the paper "<Paper Title & arXiv ID>" from scratch in MuJoCo WebAssembly and deploy a self-contained browser lab + visual essay:
1. Match the paper's exact physics timestep, actuator gains (kp/kv), torque limits, and kinematic constraints in MuJoCo XML, and run real neural inference or Web Worker iLQR in JavaScript at the paper's exact control rates.
2. Build `index.html` (Spatial UI simulator with camera presets, live telemetry plots, and disturbance injection) and `scrollytelling.html` (interactive visual essay).
3. Run a headless Chromium (Playwright) verification loop asserting 0 console errors and non-zero task progress over 300+ frames.

纯文本

已复制

借助 Argon 推进具身智能

Argon 融合了编程、视觉感知、物理仿真和具身机器人控制。通过原生推理多视角视频、3D 几何、接触力学和本体感知遥测数据，Argon 不仅能理解场景中有哪些物体，还能理解它们如何与智能体进行物理交互，以完成全身灵巧任务。Gemini 4 Argon 标志着一个重要里程碑，将加速迈向物理通用人工智能。通过构建这一核心智能，我们的目标是实现能够与人类协同工作、解决复杂挑战的机器人技术。

开始使用

你现在就可以在 Google 的开发者平台上使用 Gemini 4 Argon 开始构建：

• Google AI Studio：面向 Google AI Ultra 订阅用户提供。
• Gemini API：在 AI Studio 中通过 Gemini API 的付费层级提供。通过 gemini-4-argon 使用 Interactions API，并从第一天起支持 google-genai SDK。
• Google Antigravity：面向 Google AI Ultra 订阅用户提供。要开始构建，请在 Google Antigravity 中选择 Gemini 4 Argon（gemini-4-argon）作为当前模型，用于复杂规划、/goal 运行和多智能体编排。

上一页

下一页