google-deepmind/robotics-antigravity-plugin

i2rt-yam-ik

Solves 6-DOF inverse kinematics for the I2RT YAM and YAM Pro robot arm using Mink + DAQP on combined YAM + gripper MJCF.

소스 보기

사용 방법

  1. 설치 명령을 복사하고 자신의 에이전트 환경에 설치하기 전에 소스 패키지를 확인하세요.
  2. 필요한 도구, 인증 정보, 지원 환경은 원본 문서에서 확인하세요.
  3. 입력과 원하는 산출물을 설명하고 출처와 결과를 대조하세요.
npx skills add https://github.com/google-deepmind/robotics-antigravity-plugin --skill i2rt-yam-ik

시작 전 확인

  • 설치에는 모델 구독, API 한도, 파일 접근 권한이 포함되지 않습니다.
  • 별점 수와 설치 수는 활동 지표이며 독립적인 품질·보안 검증이 아닙니다.
원본 Skill 문서

원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.

I2RT YAM Inverse Kinematics (mink + daqp via `YamIk`)

Use bundled `YamIk` (pip install mink daqp mujoco, <1 ms/solve on Linux & macOS) parameterized by arm_type ("yam" or "yam_pro") and gripper_type ("linear_4310", "crank_4310", "linear_3507", "flexible_4310", etc.) via combine_arm_and_gripper_xml — never load standalone yam.xml when driving a gripper-equipped arm because bare yam.xml has nsite = 0:

  1. `grasp_site` Frame Convention & Jaw-Closing Axis (`±Y_site`, NOT `+X_site`):
  • YAM has 3 parallel pitch axes (J2 || J3 || J4) and a 74.0 mm J4-J5 non-spherical wrist offset.
  • In i2rt gripper MJCFs, +Z_site is the forward approach vector and parallel-jaw fingers open/close along `grasp_site` `±Y_site` (NOT `+X_site`).
  • Use ik.make_grasp_pose(pos, approach_dir, jaw_perp_axis=...) (for arbitrary 3D approach vectors — top-down, angled, or side grasps) or ik.make_topdown_grasp_pose(x, y, z, obj_yaw_rad=None, tilt_deg=0.0) (tilt_deg=10–15 near outer reach >75%):
  • Pass jaw_perp_axis=None (obj_yaw_rad=None) for symmetric objects to keep wrist roll J6 near neutral 0.
  • For anisotropic objects where jaws must close perpendicular to a target axis a, pass jaw_perp_axis=a (+X_site || ±a so ±Y_site closes across the object, automatically wrapped by ±pi using the loaded MJCF's neutral +X_site at J6 = 0 so J6 stays inside [-2.094, 2.094] rad).
  1. Reliable Seeding & Mandatory FK Verification ([yam_ik.py](scripts/yam_ik.py)):
  • Instantiate ik = YamIk(arm_type=..., gripper_type=...) once and call q_goal = ik.solve(T_target, q_seed=...).
  • Q_DEFAULT_SEED = [0.0, 1.2, 1.2, -1.2, 0.0, 0.0] avoids stalling against J4 in [-1.693, 1.571] rad on downward grasps; pass q_seed=q_actual[:6] for local moves so PostureTask regularizes smoothly around the current configuration.
  • YamIk.solve enforces solver="daqp", ConfigurationLimit, non-adjacent self-collision limits, frozen gripper DOFs (clamped to each model's jnt_range), fresh re-seeding every call, and strict MuJoCo FK verification (pos_err <= 2 mm), raising ValueError before motion if any phase waypoint is unreachable.
같은 저장소의 Skills

더 많은 Skills

모든 Skills
google-deepmind
커뮤니티

blender-creation

Creates physically calibrated, articulated, PBR-textured 3D MuJoCo assets (MJCF model.xml, .mjz, mjSpec) from scratch using headless Blender (bpy, bmesh, blendermeshutils, assetbuildutils), convex collision synthesis, 10-scenario auditing, and Studio preview. Use only when asked to create, model, or build a 3D MuJoCo object from a URL, image, or text description, or when blender-creation is named. Don't use for non-MuJoCo Blender/Cycles/CAD modeling.

1시간 설치 수
2
GitHub Stars
80
업데이트
11월 17일
google-deepmind
커뮤니티

fr3-duo-control

Franka Duo robot only (Franka Research 3 / FR3 Duo dual-arm cell with Robotiq 2F-85 grippers, ZED Mini and RealSense D405 cameras). Brings up, homes, and controls it via pylibfranka on hardware or in MuJoCo sim. Use only when the user names Franka Duo, FR3 Duo, or fr3duo.

1시간 설치 수
2
GitHub Stars
80
업데이트
11월 17일
google-deepmind
커뮤니티

i2rt-yam-bringup

Checks and installs the I2RT YAM SDK (i2rt), configures Linux socketcan or macOS gsusb CAN bringup for I2RT YAM robot arms, and provides a wave smoke test (wave.py). Use only before bringing up or programming a physical or simulated I2RT YAM robot arm, or when asked to run a wave gesture on an I2RT YAM robot. Don't use for non-YAM CAN bus debugging (e.g. automotive OBD-II or industrial PLC CAN networks) or non-robotics tasks.

1시간 설치 수
2
GitHub Stars
80
업데이트
11월 17일
google-deepmind
커뮤니티

iteration

Staged execution and video-based debugging workflow for physical or MuJoCo simulated robots: splitting a robot manipulation task into stages, testing each stage with one script, reviewing rollout videos, and verifying physical completion after release and settle. Use only when writing code that makes a physical or simulated robot act and move. Don't use for non-robotics software engineering, UI/WebGL testing, or general scripting.

1시간 설치 수
2
GitHub Stars
80
업데이트
11월 17일