Gemini 4 Argon 用于机器人

OCT. 13, 2026

Gemini 4 Argon 在物理理解、代码生成和视觉语义推理方面实现了阶跃式提升——解锁了机器人技术栈中的新工作流，并加速迈向物理 AGI 的进程。

你可以在具身推理（ERQA）基准上看到这种能力跃升，该基准测试模型在三维环境中理解和解决物理、空间问题的能力。仅凭纯推理，Argon 得分为 87%，优于其他所有前沿模型。

gemini-4-argon__benchmark__robotics

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，我们分享了 Argon 在四个关键领域的代表性提示：

机器人控制 → 通过 MuJoCo 练习实现移动操作、零样本机器人控制，以及对视频世界模型的闭环引导。
Real2Sim → 从机器人相机画面、第一人称人类视频或手绘纸面草图中重建经过物理标定的仿真孪生体。
物理理解 → 智能体式多模态后见标注，包括对灵巧性、稠密价值函数和进度估计的详细理解。
研究与自我改进 → 通过仿真改进动作，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：通过 MuJoCo 练习将物品放入垃圾桶

Argon 控制 Boston Dynamics Spot® 四足机器人（12 自由度腿部 + 6 自由度手臂）学习一项新技能：踩脚踏式垃圾桶的踏板以丢弃易拉罐。Argon 创建一个带有关节的 MuJoCo 脚踏式垃圾桶数字孪生体，在踩下实体踏板并将易拉罐丢入打开的垃圾桶之前练习其动作。

抱歉，您的浏览器不支持播放此视频

Spot 移动操作 — 从真实到仿真再到真实的脚踏式垃圾桶丢弃：Spot 捡起易拉罐，检查其腕部相机，在 MuJoCo 中练习三腿三脚架平衡，踩下脚踏板，验证手在打开的垃圾桶中的居中，并将易拉罐丢入其中。

提示：

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，将倒下的杯子扶正，并把揉皱的餐巾纸放进杯子里。Argon 通过视觉坐标框架和动作基元进行推理，在多个回合中完成该任务。

抱歉，您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：经过标定的腕部相机，带有投影的 2 厘米世界坐标和夹爪闭合准星。

提示：

[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. 全控

Argon 在 Gemini Omni 世界模型中操控机器人，将五块散落的 Scrabble 字母牌（G, R, A, N, O）重新排列成整齐的一行，拼出 A - R - G - O - N。Argon 不是指定一个静态提示，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，检查中间展开关键帧，并优化其提示。

抱歉，您的浏览器不支持播放此视频

迭代视频世界模型操控灵巧手：33 步受控视频展开拼出 A-R-G-O-N。

提示：

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. 机器人视角到仿真

Argon 查看真实机器人遥操作片段并开发数字孪生；它获取或创建仿真资产，使其在与灵巧手接触时看起来和表现得像真实物体。Argon 在迭代视觉推理循环中调整资产和场景，使仿真视觉效果和接触事件与真实录制数据中的相匹配。

抱歉，您的浏览器不支持播放此视频

左侧（真实视图）：同步的真实机器人相机画面；右侧（模拟）：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. 人类视角到模拟

给定一段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，用复合凸分解网格替代粗糙的边界框以表示凹容器，并将人类手部轨迹重定向到 MuJoCo 接触物理中的 76 自由度双臂人形机器人，其双手为 22 自由度五指手。

抱歉，您的浏览器不支持播放此视频

人类视角到仿真——从第一人称人类视频到灵巧人形孪生：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 多视角仿真推演。

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. 草图到仿真

使用 Argon，开发者可以拍摄一张粗略的纸笔草图——草图中显示一个 Apollo 2 人形机器人沿抛物线弧投掷一个球，以推倒桌上 3-2-1 堆叠的箱子——然后让 Argon 自动构建 GPU 加速的 mujoco_warp 场景，合成平衡的 76 自由度全身及五指投掷轨迹，在闭环物理中验证 6/6 目标击倒，并导出 Blender 4.5 Cycles 渲染。

抱歉，您的浏览器不支持播放此视频

从草图到仿真——手绘草图到灵巧人形轨迹 & Cycles 渲染。

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

使用 Argon 解锁物理理解

1. 后见标注与数据质量

Argon 对灵巧操作具有精确的物理理解，并能够从视频中提供详细的分割和文本标注。Argon 能够识别较差的操作片段，例如停顿、不稳定抓取和失误。给定一段俯视视频，其中双臂 ALOHA 2 机器人在 136 秒内将一张方形纸折叠成折纸狐狸，Argon 生成时间对齐的 4 级后见标注层级：回合目标（G1）、6 个连续的语义子任务（G2）、17 个接触级原子原语（G3），以及自动质量控制（QC）重加权，该重加权在失误和重试期间保持任务进度不变。

抱歉，你的浏览器不支持播放此视频

Argon 多粒度后见 & 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. 找出异常球

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，其中包含 12 个外观相同的台球（其中一个秘密地更重或更轻）和一个铰接式 Roberval 双盘天平。随后，它编写了一个仅基于视觉的闭环求解器，该求解器通过立体 RGB 相机对球进行三角测量，在每次称重后视觉读取天平倾斜，并在 3 次称重内将异常球隔离到重（红色）或轻（蓝色）托盘中。

抱歉，您的浏览器不支持播放此视频

找出异常球（仅视觉闭环求解器）：双臂 Franka FR3 在金色/青色托盘上称量候选球，读取立体秤倾斜，并在 3 次称量中将确认的重异常球隔离到红色托盘中。

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 是一个用于机器人任务的通用价值函数。在来自人类 UMI 数据集的一段 60 秒双手戒指分拣回合中评估，Argon 围绕接触转换迭代检查同步的头部、左夹爪和右夹爪相机画面，以生成密集的价值函数 / 任务进度预测，并将轨迹分解为 28 个具有正、中、负进度斜率的片段。

抱歉，您的浏览器不支持播放此视频

价值函数：同步的三相机 UMI 播放（头部、左夹爪、右夹爪），与动画任务进度曲线和双手片段标注锁定。

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

使用 Argon 解锁研究与自我改进

1. 通过仿真改进动作

Argon 可以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。从一张真实桌面的 RGB 照片出发，Argon 自主构建一个经过标定的 MuJoCo 仿真孪生体，定义新任务，并合成一个闭环运动生成器，以在仿真中收集 4,500+ 个域随机化回合——将真实世界物理机器人的成功率从 56% 提升至 75%，并将完成时间从 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：提升真实世界评估基准（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. 视觉拾取与放置的自改进

在一条简单的 6-DoF 机械臂上，Argon 通过试验和自我反思，在拾取与放置任务中迭代地自我改进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 视觉引导拾取与放置 — 跨迭代 1 → 2 → 3 的硬件迭代自改进：迭代 1 (58.9s)，迭代 2 (22.9s，放置时擦到盒子边缘)，迭代 3 (17.4s，精准居中放置)。

提示词：

# 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. 单提示词在 MuJoCo WebAssembly 中复现机器人学论文

每篇论文只需一个提示词，Argon 即可读取机器人控制手稿及任何已发布的检查点，将物理模型和闭环控制器（从 DeepLoco 中的 7 网络分层运动，到 HUSKY 中的 50 Hz PPO 执行器，再到 Whole-Body MPC 中的 Web Worker iLQR）移植到自包含的 MuJoCo WebAssembly 浏览器实验环境中，并在无头 Chromium 中对其进行验证。

抱歉，您的浏览器不支持播放此视频

五篇里程碑式的腿足 & 类人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、全身 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 桥接编程、视觉感知、物理模拟和具身机器人控制。通过对多视角视频、3D 几何、接触力学和本体感觉遥测进行原生推理，Argon 不仅理解场景中存在哪些物体，还理解它们如何与智能体发生物理交互，以完成全身灵巧任务。Gemini 4 Argon 标志着加速通往物理 AGI 之路的重要里程碑。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。

快速入门

如今，你可以在 Google 的各个开发者平台上开始使用 Gemini 4 Argon 进行构建：

Google AI Studio: 可供 Google AI Ultra 订阅者使用。
Gemini API：在 AI Studio 中，Gemini API 的付费层级可用。在 Interactions API 中使用 gemini-4-argon，并在 google-genai SDK 中提供首日支持。
Google Antigravity：面向 Google AI Ultra 订阅者。要开始构建，请在 Google Antigravity 中选择 Gemini 4 Argon（gemini-4-argon）作为您的当前模型，用于复杂规划、/goal 运行和多智能体编排。