Aerial World Model for Long-horizon Visual Generation and Navigation in 3D Space

TL;DR

ANWM presents a novel aerial navigation world model that generates long-horizon, first-person visual observations conditioned on past frames and 3D actions. The key innovation is the Future Frame Projection, which provides a coarse geometric prior to stabilize long-range generation, combined with Independent Latent Modulation and a CDiT-based backbone to produce temporally consistent visuals. The authors release a large-scale dataset (approximately 350k training segments) and show that ANWM outperforms 2D/indoor baselines on both visual generation (up to 32 s horizons) and navigation metrics, while providing interpretable trajectory ranking via perceptual similarity to a target. Limitations include mode collapse for very long ranges and texture distortions, with future work aimed at stronger physical constraints and active planning integration to close the loop between imagination and real navigation.

Abstract

Unmanned aerial vehicles (UAVs) have emerged as powerful embodied agents. One of the core abilities is autonomous navigation in large-scale three-dimensional environments. Existing navigation policies, however, are typically optimized for low-level objectives such as obstacle avoidance and trajectory smoothness, lacking the ability to incorporate high-level semantics into planning. To bridge this gap, we propose ANWM, an aerial navigation world model that predicts future visual observations conditioned on past frames and actions, thereby enabling agents to rank candidate trajectories by their semantic plausibility and navigational utility. ANWM is trained on 4-DoF UAV trajectories and introduces a physics-inspired module: Future Frame Projection (FFP), which projects past frames into future viewpoints to provide coarse geometric priors. This module mitigates representational uncertainty in long-distance visual generation and captures the mapping between 3D trajectories and egocentric observations. Empirical results demonstrate that ANWM significantly outperforms existing world models in long-distance visual forecasting and improves UAV navigation success rates in large-scale environments.

Figures (11)

• Figure 1: Visual navigation in large-scale aerial space. Given a visual target, the agent is required to plan a trajectory whose final observation aligns with the target. We leverage a world model that imagines visual observations along all possible trajectories. By computing the similarity between the imagined observations and the target, the optimal trajectory is determined. This imagination-based planning paradigm potentially reduces the navigation cost in large-scale open 3D environments.
• Figure 2: The Framework Overview. a) We collect the datasets from AVLN simulators and generate trajectory clips by action enrichment and random partition. b) For single-frame generation, ANWM produces future visual observations conditioned on the noisy latent, the past-frame latent, the projected future-frame latent, and the embedding of the upcoming action. We employ the Future Frame Projection module to warp the past frame into the future viewpoint, providing a strong scene prior for generation. c) For long-horizon generation, ANWM operates in an autoregressive manner to generate sequential visual observations along the trajectory. Each newly generated frame is appended to the past-frame queue which is then used as input for the next observation generation.
• Figure 3: Model Architecture. ANWM adopts CDiT bar2025navigation as the backbone but uses the past frame and the projected future frame as distinct conditional signals to control the generation process. Specifically, ANWM first splits the condition latents into the past-frame latent and the projected future-frame latent, and applies separate scale and shift parameters to modulate the strength of the conditioning signal. The modulated latents are then fed into two shared-weight Multi-Head Cross-Attention branches.
• Figure 4: The qualitative results of generative visual observation along the path. Left: 2D trajectory. Right: 3D trajectory.
• Figure 5: The qualitative results of visual navigation. ANWM ranks each trajectory's final prediction by measuring the LPIPS similarity with the goal Image. The trajectory with the lowest LPIPS is selected for execution. We only visualize the top-3 trajectories.