SkyJEPA learns long-horizon latent dynamics for quadrotors via JEPA plus a physics prober, enabling zero-shot sim-to-real control with sampling-based MPC and automated sim data generation.
Skydreamer: Interpretable end-to-end vision-based drone racing with model-based reinforcement learning,
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
fields
cs.RO 4years
2026 4verdicts
UNVERDICTED 4representative citing papers
MAD learns recurrent latent dynamics to reconstruct robocentric occupancy and visibility grids, yielding higher success rates and faster flight than vision-only baselines in simulation and real-world quadrotor experiments.
Robustness of world models during cross-environment SSL pretraining predicts sim-to-real transfer success for quadrotor navigation, with discrete latent size and training sequence length as dominant factors.
AirDreamer combines world-model-based environment understanding with an RL policy and sparse rewards to navigate unseen environments, achieving 5.3% higher success than baselines and effective sim-to-real transfer without tuning.
citing papers explorer
-
SkyJEPA: Learning Long-Horizon World Models for Zero-Shot Sim-to-Real Control of Quadrotors
SkyJEPA learns long-horizon latent dynamics for quadrotors via JEPA plus a physics prober, enabling zero-shot sim-to-real control with sampling-based MPC and automated sim data generation.
-
MAD: Mapping-Aware World Models for Agile Quadrotor Flight
MAD learns recurrent latent dynamics to reconstruct robocentric occupancy and visibility grids, yielding higher success rates and faster flight than vision-only baselines in simulation and real-world quadrotor experiments.
-
Generalization of World Models under Environmental Variability for Vision-based Quadrotor Navigation
Robustness of world models during cross-environment SSL pretraining predicts sim-to-real transfer success for quadrotor navigation, with discrete latent size and training sequence length as dominant factors.
-
AirDreamer: Generalist Drone Navigation with World Models
AirDreamer combines world-model-based environment understanding with an RL policy and sparse rewards to navigate unseen environments, achieving 5.3% higher success than baselines and effective sim-to-real transfer without tuning.