arXiv:2608. 03244v1 Announce Type: new Abstract: Image-goal visual navigation is a fundamental capability for embodied agents.
Image-goal visual navigation is a fundamental capability for embodied agents. Existing navigation policies efficiently predict waypoint trajectories but lack visual foresight, while navigation world models can anticipate future observations but often require costly planning rollouts.
DiffWAM is a geometry‑conditioned navigation world‑action model that transforms predictive features from a frozen video foundation model into continuous camera trajectories, eliminating the need for future‑video synthesis and multi‑frame reconstruction during deployment. Its Grid‑Motion module preserves spatial‑temporal motion associations, while Latent2Pose grounds them with first‑frame geometry to recover metrically meaningful 3D motion. The system, complemented by FastDreamer for asynchronous trajectory handoff, achieves a trajectory RMSE of 0.3492 m and a 74.40 % endpoint success rate on the DiffWAM‑1000 benchmark, with real‑world tests showing complex UAV behaviors and an onboard implementation reaching 1.08 s latency on NVIDIA Jetson AGX Thor.
WALT introduces a method to align latent trajectories with pretrained driving world models, creating a compact generative trajectory space that preserves action-relevant semantics without altering the original model. The approach uses a dual-branch autoencoder to map raw waypoints into this latent space and transfers visual world knowledge into trajectory representations. Experiments on NAVSIM benchmarks show modest performance gains and a 30.5% reduction in planner FLOPs, indicating that maintaining world representations while extracting action-relevant information can improve trajectory planning efficiency.
arXiv:2606. 29908v1 Announce Type: cross Abstract: Existing world model-based planners for visual navigation typically follow a verification-centric paradigm, decoupling goal intent from trajectory synthesis.
The paper introduces Spatially Aware World Action Model (SA‑WAM), a diffusion‑based framework that extends existing World Action Models by incorporating depth information alongside RGB to enable 3‑D‑aware action and future‑state prediction. SA‑WAM repurposes a pretrained video diffusion model, using a nonlinear encoding to map unbounded depth into the tokenizer’s bounded domain, thus preserving pretrained visual priors without 3‑D‑specific fine‑tuning. The model achieves state‑of‑the‑art performance on RoboCasa and LIBERO‑Plus benchmarks and demonstrates superior real‑world performance on a UR5 robotic arm in randomized environments, while also providing analysis linking world‑model prediction quality to rollout success.
arXiv:2607. 14739v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have achieved impressive results in visuomotor policy learning, yet remain fundamentally reactive, mapping current observations and language to actions without explicit forward prediction of world dynamics.
arXiv:2603. 25937v2 Announce Type: replace-cross Abstract: Visual Navigation Models (VNMs) promise generalizable, robot navigation by learning from large-scale visual demonstrations.
arXiv:2606. 17294v1 Announce Type: cross Abstract: Vision Navigation Foundation Models (VNMs) promise end-to-end learned navigation policies capable of zero-shot deployment across diverse embodiments and environments.
Conventional visual navigation policies often struggle with myopic decision-making and mode collapse in complex environments. While world models offer a promising alternative, existing paradigms typically isolate perception, generation, and control, failing to capture their shared spatio-temporal dynamics.
arXiv:2606. 06147v1 Announce Type: new Abstract: End-to-end Vision-Language-Action (VLA) models have shown promise in UAV navigation.
arXiv:2606. 07244v1 Announce Type: cross Abstract: Vision-Language Navigation in Continuous Environments (VLN-CE) requires agents to follow natural-language instructions while navigating in real-world-like environments.
GeoWAM introduces a visual geometry world action model that predicts future scene geometry instead of future images, using point clouds to capture spatial structure and transformations. The model is pretrained to forecast geometry, then a geometry-conditioned action head predicts ego trajectories. Experiments show that this geometry-based approach yields stronger driving policies than image-based alternatives.