LeCropFollow enables zero-shot sim-to-real visual navigation in unstructured crop fields by optimizing trajectories in a latent manifold from self-supervised semantic heatmaps with TD-MPC2, reducing semantic failures 2.4x versus keypoint baselines in corn gaps.
Dyna, an integrated architecture for learning, planning, and reacting
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A K-fold cross-fitted proximal bridge estimator for reward-emission and observation-transition functions in confounded POMDPs, with an oracle-comparator error bound decomposed into nuisance and averaging terms.
citing papers explorer
-
LeCropFollow: Latent Space Planning for Navigation in Unstructured Crop Fields
LeCropFollow enables zero-shot sim-to-real visual navigation in unstructured crop fields by optimizing trajectories in a latent manifold from self-supervised semantic heatmaps with TD-MPC2, reducing semantic failures 2.4x versus keypoint baselines in corn gaps.
-
Cross-fitted Proximal Learning for Model-Based Reinforcement Learning
A K-fold cross-fitted proximal bridge estimator for reward-emission and observation-transition functions in confounded POMDPs, with an oracle-comparator error bound decomposed into nuisance and averaging terms.