Part of Ethgar's robot lab notebook: the 5k reference model of a clean-room SkyJEPA reimplementation (5.6k parameters; flies circle, figure-8 and lemniscate by MPPI through the learned model in sim). The code repository is private for now.

SkyJEPA — quadrotor latent-dynamics world model

A clean-room PyTorch reimplementation of SkyJEPA (Rao et al., 2026, arXiv:2606.23444): a JEPA-style latent-dynamics world model for quadrotors, trained entirely on domain-randomized simulation, with a physics-inspired prober and an MPPI controller. Trained with the full profile.

• Training data: Ethgar/skyjepa-quadrotor-sim
• Method: two causal-TCN encoders (state + action) → GRU latent predictor, anti-collapse via SIGReg (LeJEPA), a frozen-backbone physics prober that outputs residual accelerations correcting a differentiable SO(3) rigid-body integrator. Encoder+predictor stack is ~5.6k parameters.
• Inputs: 18-dim state (position, velocity, rotation matrix, body rates) and 4-dim action histories at 20 Hz. State-based (GPS/IMU), not vision.

Files

file	contents
jepa.pt	stage-1 encoders + GRU predictor + normalizer
prober.pt	stage-2 physics prober (frozen backbone)
baseline.pt	predictive-MLP baseline (optional)
export/skyjepa_stack.ts	TorchScript deploy stack (optional)
export/skyjepa_stack.onnx	ONNX deploy stack (optional)

Usage

pip install "skyjepa[hf] @ git+https://github.com/edgarmoreaualix/LeDrone.git#subdirectory=skyjepa"
huggingface-cli download Ethgar/skyjepa --local-dir checkpoints/full
python scripts/run_mppi_demo.py --profile full --ckpt-dir checkpoints/full

License & credit

MIT. Method credit belongs to the original SkyJEPA authors (Rao et al., 2026, arXiv:2606.23444); this is an independent clean-room reimplementation, not affiliated with or endorsed by them.