README.md

# vfm-dexmimicgen

# Rebase to IsaacLab External 

[Updated 07/22/2026]

## Installation

1. Install [IsaacLab](https://isaac-sim.github.io/IsaacLab/release/3.0.0-beta2/index.html) 's `release 3.0.0-beta2`

Activate your `env_isaaclab`
```bash
# e.g.
source <your_path_to>/env_isaaclab/bin/activate

## e.g. 
# source /home/yizhou/Projects/IsaacLab/env_isaaclab/bin/activate

```

2. Install repo

```bash

# install
python -m pip install -e .

## or 
# uv pip install -e .
```

## Usage

1. List env
```bash
python scripts/list_envs.py
```

2.zero-agent
```bash
python scripts/zero_agent.py --task Template-Sdg-Mimic-Gen-v0 --viz kit
```

3. link asset
```bash
ln -s /path/to/dexmimicgen_custom_assets ./custom_assets

##  e.g. 
# ln -s /home/yizhou/Downloads/dexmimicgen_custom_assets ./custom_assets
```

## Teleop

Install [Isaac Teleop]() under any python env

```bash
# From PyPI
pip install 'isaacteleop[cloudxr,retargeters]~=1.3.131' --extra-index-url https://pypi.nvidia.com
```

Start
```bash
python -m isaacteleop.cloudxr --accept-eula
```

Open the [Web Client](https://nvidia.github.io/IsaacTeleop/client/v1.3.131/#/sim)


Activate CloudXR environment in another terminal in your IsaacLab env: 
```bash
source ~/.cloudxr/run/cloudxr.env
## e.g.
# source /home/yizhou/.cloudxr/run/cloudxr.env
```


```bash
python scripts/teleop_se3_agent_bimanual_xr.py \
  --task Template-YAM-Play-v0 \
  --teleop_device motion_controllers \
  --enable_cameras \
  --viz kit \
  --num_envs 1 \
  --enable_gripper \
  --xr 

# python scripts/teleop_se3_agent_bimanual_xr.py \
#   --task Template-UR10-Play-v0 \
#   --teleop_device motion_controllers \
#   --enable_cameras \
#   --viz kit \
#   --num_envs 1 \
#   --xr \
#   --reverse_rotation_yz
```

---

# YAM bimanual task suite (30 tasks)

Scripted bimanual manipulation tasks for the YAM arm in Isaac Lab, laid out the way ManiSkill
lays out its tasks: one registered class per task file, with the environment, the solvers and the
motion planner as separate layers. Built on RoboTwin 2.0 assets. Every task ships a scripted
solver, so it generates demonstrations *and* a pass/fail signal — not a policy.

![task overview](outputs/frames/overview.png)

```bash
export ROBOTWIN_USD=./robotwin_usd
export OMNI_KIT_ACCEPT_EULA=YES

python scripts/yam_task.py --list
python scripts/yam_task.py --task grape_box --seed 3 --video outputs/tasks/grape_box.mp4 \
    --kit_args="--/rtx/verifyDriverVersion/enabled=false"

python scripts/generate.py --tasks passing --episodes 100    # scale up to a dataset
```

```
source/bimanual/yam/
  motion/     how the robot MOVES      arm.py · planner.py · recorder.py
  envs/       the world a task runs in  base_env.py · scene.py · config.py
  solvers/    scripted skills           pick_place · multi_pick · insert · stack · dual_lift ·
                                        push · handover · tool_use · pour · pull · sort ·
                                        shelf · articulate · rope
  tasks/      ONE FILE PER TASK, registered by name;  configs/ one YAML each
  registry.py · conditions.py
scripts/      runners · asset converters · generation · verification
```

## Documentation

| | |
|---|---|
| [`doc/capabilities.md`](doc/capabilities.md) | what the suite supports — skill families, randomization, articulations |
| [`doc/setup.md`](doc/setup.md) | environment, configuration, running a task |
| [`doc/generation.md`](doc/generation.md) | scaling up to a demonstration dataset |
| [`doc/tasks.md`](doc/tasks.md) | all 30 tasks, current pass/fail, and how they differ from RoboTwin's |
| [`doc/assets.md`](doc/assets.md) | asset conversion pipeline and licence |
| [`doc/verification.md`](doc/verification.md) | agent-in-the-loop visual verification |
| [`DIAGNOSTICS.md`](DIAGNOSTICS.md) | **every non-obvious failure and the measurement that settled it — read this before debugging a new asset** |
| [`source/bimanual/yam/README.md`](source/bimanual/yam/README.md) | architecture and how to write a task |