arXiv AI
GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Jun 5
3D Underwater Path Planning via Generative Flow Field Surrogates
arXiv:2606. 06077v1 Announce Type: cross Abstract: Autonomous underwater vehicle (AUV) launch and recovery (LAR) into the hull of an advancing host platform requires traversal of a complex, three-dimensional propeller wake whose hydrodynamic structure cannot be characterised by a uniform current model.
arXiv Machine Learning
Jul 1
Pretrained Video Models as Differentiable Physics Simulators for Urban Wind Flows
arXiv:2603. 21210v3 Announce Type: replace Abstract: Designing urban spaces that provide pedestrian wind comfort and safety requires time-resolved Computational Fluid Dynamics (CFD) simulations, but their current computational cost makes extensive design exploration impractical.
arXiv AI
Jul 10
GenDA: Generative Data Assimilation on Complex Urban Areas via Classifier-Free Diffusion Guidance
arXiv:2601. 11440v3 Announce Type: replace-cross Abstract: Urban wind flow reconstruction is essential for assessing air quality, heat dispersion, and pedestrian comfort, yet remains challenging when only sparse sensor data are available.
arXiv Machine Learning
Jun 25
AeroCast: Probabilistic 3D Trajectory Prediction for Non-Cooperative Aerial Obstacles via Transformer-MDN Architecture
arXiv:2606. 25122v1 Announce Type: cross Abstract: Autonomous aerial vehicles operating in shared airspace must predict the future positions of non-cooperative obstacles to plan evasive maneuvers before a collision becomes unavoidable.
arXiv Machine Learning
Jul 22
Reinforcement Learning for Delivery Drone-Based Participatory Sensing in Dynamic Environments
arXiv:2607. 18874v1 Announce Type: new Abstract: Using Unmanned Aerial Vehicle (UAV) for urban sensing has emerged as a powerful paradigm to monitor the status of the city, e.
arXiv AI
3d ago
DiffWAM: A Fast and Efficient Navigation World Action Model
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.