{
  "name": "PhaForce",
  "url": [
    "https://huggingface.co/datasets/wangmingxinthu/PhaForce-Dataset"
  ],
  "version": "1.0.0",
  "description": "Real-robot contact-rich manipulation dataset from PhaForce, converted to FTP-1/OXT-style zarr keys with visual observations, robot state, gripper state, and 6D FlexivGripperTorque wrench state.",
  "type": "gripper",
  "sensor_image": [],
  "sensor_array": [],
  "sensor_state": [
    "FlexivGripperTorque"
  ],
  "task": 4,
  "trajectory": 260,
  "frame": 116870,
  "contact": [
    "https://thu-wangmx.github.io/"
  ],
  "reference": {
    "paper": "PhaForce: Phase-Scheduled Visual-Force Policy Learning with Slow Planning and Fast Correction for Contact-Rich Manipulation",
    "website": "https://thu-wangmx.github.io/phaforce/",
    "authors": "Mingxin Wang, Zhirun Yue, Renhao Lu, Yizhe Li, Zihan Wang, Guoping Pan, Kangkang Dong, Jun Cheng, Yi Cheng, Houde Liu."
  },
  "code_trigger": "PhaForce_FTP1_Converted"
}