dontKnow23456/rambo-go2-policies

RAMBO Go2 policies

Final policy checkpoints produced by reproducing RAMBO: RL-augmented Model-based Optimal Control for Whole-body Loco-manipulation.

Included checkpoints

Task	Checkpoint	Training configuration
Isaac-RAMBO-Quadruped-Go2-v0	quadruped/model_2000.pt	4096 environments, 2000 PPO iterations, seed 42
Isaac-RAMBO-Biped-Go2-v0	biped/model_4000.pt	4096 environments, 4000 PPO iterations, seed 42

Each model_*.pt is a CRL2 PyTorch training checkpoint. It contains the policy, value network, optimizer state, observation normalizer, and completed iteration count. It is not a Transformers model or an inference-only safetensors file.

The accompanying agent.yaml and env.yaml files record the training configuration. The checkpoint paths intentionally exclude training videos, TensorBoard events, W&B files, and Python pickle configuration files.

SHA-256 checksums:

quadruped/model_2000.pt  1cc5f68fe15e37ccabae26060d79a26a8c078ed465a81f6f729c2009b67ca706
biped/model_4000.pt     c16e64bf1ca2dc16878c386b742cd303e65040f52e8744cd0c96c540c595b2a6

Use with RAMBO

Use the matching RAMBO/Isaac Lab environment and place a downloaded checkpoint under the corresponding CRL2 log root before launching play.py. For example, for the Quadruped policy:

hf download dontKnow23456/rambo-go2-policies quadruped/model_2000.pt \
  --local-dir /tmp/rambo-go2-policies

mkdir -p logs/crl2/rambo_quadruped/hf_quadruped/params
cp /tmp/rambo-go2-policies/quadruped/model_2000.pt \
  logs/crl2/rambo_quadruped/hf_quadruped/

python scripts/reinforcement_learning/crl2/play.py \
  --task Isaac-RAMBO-Quadruped-Go2-v0 \
  --num_envs 1 \
  --load_run hf_quadruped \
  --load_checkpoint 2000

For Biped, substitute task Isaac-RAMBO-Biped-Go2-v0, log root rambo_biped, run name hf_biped, and checkpoint 4000.

These checkpoints were generated from source commit 6f6c87c614255a5ada063da40206a2fa67054cb6 of catachiii/rambo, using Isaac Sim 4.5.0, Isaac Lab, CRL2, and PyTorch. The local reproduction includes checkpoint resume/normalizer-state fixes; use the accompanying workspace source for the most reliable loading behavior.

env.yaml is an Isaac Lab configuration dump that contains NumPy-specific YAML tags. Treat it as trusted project configuration and load it only within a trusted RAMBO/Isaac Lab environment.

License and citation

The source RAMBO repository is licensed under CC BY-NC 4.0. These artifacts are shared under the same non-commercial terms; see LICENSE-CC-BY-NC-4.0.md.

If you use RAMBO, please cite:

@article{cheng2025rambo,
  title={RAMBO: RL-augmented Model-based Optimal Control for Whole-body Loco-manipulation},
  author={Cheng, Jin and Kang, Dongho and Fadini, Gabriele and Shi, Guanya and Coros, Stelian},
  journal={arXiv preprint arXiv:2504.06662},
  year={2025}
}

Safety

PyTorch .pt files are pickle-based checkpoints. Only load files obtained from trusted sources, and use them in simulation before considering any hardware deployment.